1920 metalsinaircraft00hanbrich

8/6/2019 1920 metalsinaircraft00hanbrich

1/128

715

ETALSIN AIRCRAFTCONSTRUCTION

COCOo

WILFRED HANBY


2/128

SPECIALHIGH GRADE ALLOY

3FOAIMOIL HAIan N -, yV

/YV//3

VICKERSLIMITEDii.ilT ;>

"DURALUMIN."(Regd. Trade Mark)

Light as Aluminium. Strong as Steel.Specific Gravity. 2'8. Tensile Strength up to 35 tons'

An alloy of Aluminium with the strength and hardness of Mild Steel,but having only one-third of its weight. Will not rust. Almostunaffected by sea^water.

"DURALUMIN"

It is anjideal substitute for Aluminium, Nickel, Silver, Brass, Copper,Nickel-plated and Silvered articles, and it is the only substitute forSteel where lightness combined with the strength of that metal isrequired. For Scientific Apparatus, Instruments of Precision, Ortho-

paedic Apparatus and Hospital Supplies, etc.For Aeroplanes, Dirigibles, Hydroplanes, Motor Car, Yacht andBoat Fittings, etc.

VICKERS HOUSE, BROADWAY, LONDON, S.W.ILicensee and Distributor for the United States :THE AMERICAN DURALUMIN CO.,

Hanover Bank Building, Nassau and Pine Streets, New York.


11/128

FOREWORDIn November, 1918, when the Armistice was signed, which

brought to an end the European War, it became evident that manyyoung men would require intensive technical training in order tocomplete the engineering education they had commenced, andwhich, at the Nation's call they were, for the time being, forcedto abandon. At this juncture the Aeronautical Institute of GreatBritain undertook the establishment of courses in AeronauticalEngineering specially designed to meet the needs of a number ofthem. One of such courses the fourth one was held for thebenefit of members of the New Zealand Expeditionary Force, and,in connection with this course, Mr. Wilfred Hanby, a recognizedmaster of his subject, was good enough to give, in the summer of1919, a series of lectures, on " Aircraft Metals," which form thebasis of the present volume.The chief purpose of the lectures was to describe, in broad out-

lines, the testing, the treatment and the application of the specialalloy steels used in the construction of aircraft and their motors.

It may, perhaps, be thought that, for the construction of air-craft, there is little need to study metallurgy and its applications.It is true that, without any knowledge of metallurgical principles,one may design and build an apparently efficient and reliable-looking aeroplane ; but, when the machine is put to the test, thequestions always arise whether the right materials have been usedin the right places, and whether the materials employed were ofthe best quality obtainable.During the tests the crankshaft of the engine may take apermanent twist with the torque, or it may even suddenly fracture ;a bolt may break off ; a bearing may wear away too quickly ; aRAF-wire or a turnbuckle may snap, when least expected. Why

is it that these parts have failed in service? Simply because theyeither contained original defects, or because the materials of whichthey were composed were not in the proper physical state to with-stand the conditions of service.For such reasons a knowledge of the principles of metallurgy

will stand the aircraft constructor in good stead. It will indicateto him what metal to use for a given part ; the physical condition


12/128

in which the metal should be in order to fulfil the particular servicerequired of the part ; and also how to secure the required conditionbefore the part is put into actual service.Although the design and construction of aircraft are now well

past the experimental stage, and the machines of to-day aremanufactured on sound mechanical principles, nevertheless, thesuccess of commercial aviation will increasingly depend onreliability of construction ; and provision must always be madefor unforeseen emergencies by having a sufficient reserve ofstrength in the various members of the aircraft structure. It ishighly desirable, therefore, that aeronautical engineers shouldhave both a thorough knowledge of the treatment and strengthof the various materials used for constructional purposes, and alsoa sound understanding of how best to apply those materials tospecific parts.

It is realized that the present volume does not cover, eitherpractically or scientifically, the entire field of special alloy steelsbut the book will be found to contain much information of greatvalue to the aeronautical and motor engineer.

Mr. Hanby's clear exposition of the subject cannot fail to giveto all those connected with the design, construction and manu-facture of aircraft, a good understanding or an enlarged oneof the best way to fulfil the demands of the various aircraft materialspecifications, and it will, at the same time, guide them to anappreciation of the capabilities of alloy steels and the wideness oftheir application.

L. BLIN DESBLEDS.The Aeronautical Institute of Great Britain.

London.


13/128

AUTHOR'S PREFACETHIS book is the outcome of a series of lectures given by theauthor in connection with one of the Aeronautical EngineeringCourses of the Aeronautical Institute. It contains the substanceof the lectures as delivered, supplemented by additional matter.

The Aeronautical Engineer a term which, in the broader sense,includes the Motor Engineer is called upon to deal with a largevariety of complex metals; and it is only by correctly manipulat-ing these metals, in the various processes of manufacture, that hecan be sure of obtaining, in his finished products, the maximumpossible efficiency.

This volume explains the main principles involved in the work-ing up aircraft metals, and gives, finally, the physical propertiesand precise use of the individual steels, after correct treatment.

W. H.Rotherham.


14/128

HADFIELDS LTDWorks area200 acres.

East Hecla and Hecla Works, workmen employedSHEFFIELD, ENGLANDMANUFACTURERS OF SPECIAL ALLOYHIGH TENSILE STEELS FOR AIRCRAFTAND MOTOR CAR ENGINES"HECLA-' AND "ERA"Trade Mark BRANDS Trade Mark

HADFIELDS 17 SHEFFIELD

Hadfield's special steel six-throw Crank Shafts for Airplane Engines.Manufacturers of"HECLA" STEEL FORCINGS

(Trade Mark)For Marine and other Engineering purposes in the rough, rough machined or finished.STEEL CASTINGS

of every description up to 30 tons in weightSOLE MAKERS'OF HADFIELD'S PATENTMANGANESE STEEL

The Supreme Material for the Wearing Parts of StoneBreaking & Ore Crushing Machinery, Dredger Parts, etc.

(Trade Mark)


15/128

CONTENTSPAGEFOREWORD - 5

AUTHOR'S PREFACE ||j ' 7CHAPTER I

INTRODUCTION - 15THE WORK OF THE METALLURGIST THE SELECTING AND HANDLINGOF METALS SPECIAL STEELS TERNARY AND QUARTENARY STEELSADVANTAGES OF SPECIAL STEELS THE PRODUCTION OF HIGHTENSILE ALLOY STEELS FOR AIRCRAFT SCIENTIFIC HEAT TREAT-MENT OF STEEL.

CHAPTER IITESTING THE STRENGTH OF MATERIALS - 18

SPECIAL IMPORTANCE OF STRENGTH OF MATERIALS TO AIRCRAFTDESIGNERS STRESSES INDUCED IN AIRCRAFT PARTS VARIATIONIN THE PROPERTIES OF MATERIALS AND THE IMPORTANCE OF TESTINGTHE VARIOUS TESTS APPLIED TO AIRCRAFT STEELS : HARDNESS,TENSION, IMPACT, BENDING, FATIGUE HARDNESS TESTINGBRINELL BALL MACHINE AND HARDNESS NUMBER THE AVERYBRINELL HARDNESS TESTING MACHINE SHORE SCLEROSCOPETENSILE TESTS : ELASTIC LIMIT, YIELD POINT, MAXIMUMSTRESS, EXTENSIBILITY, REDUCTION OF AREA STANDARD FORM OFTEST-PIECE IMPACT TESTS THE IZOD MACHINE VALUE OFIMPACT TESTS HEAT TREATMENT AND IZOD RESULTS THECHARPY IMPACT TESTING MACHINE IMPACT FATIGUE TESTS : THESTANTON AND EDEN-FOSTER MACHINES BENDING TESTS.

CHAPTER IIIDEFECTS IN STEEL - 40BASIS FOR PROPER SELECTION OF MATERIAL WORKSHOP DIFFI-

CULTIES CONCERNING MATERIALS FOR SPECIAL PURPOSES ELIMINA-TION OF SULPHUR AND PHOSPHORUS OCCLUDED GASES ELIMINA-TION OF DEFECTS IN STEEL INGOTS : PIPING, BLOW-HOLES, ROAKS,LAPS.

CHAPTER IVTHERMAL AND MECHANICAL TREATMENT OF STEEL 47HEATING PROCESSES FORGING OPERATIONS THE "BURNING" OFSTEEL ANNEALING NORMALIZING HARDENING TEMPERING.


16/128

TRADE MARK

ALLOY& CARBONSTEELSFOR

AIRCRAFT 6 AUTOMOBILECONSTRUCTION

NICKEL, NICKEL CHROME,CHROME VANADIUM, SPECIAL BALLRACE STEELS inINGOTS, BILLETS, BARS or FORCINGS

OUR SERVICES are always atthe command of customers. Letus give you the benefit of ourexperience.

SirW. G. Armstrong,Whitworth & C LCENTRAL COMMERCIAL DEPT.,8, GT GEORGE STREET, WESTMINSTER, LONDON, S.W. 1.

td,

Telephone: Victoria 4010. Telegrams: "Arwhitcoy, Parl, London."DISTRICT OFFICES:BIRMINGHAM: 158-161, fit. Charles Street. MANCHESTER: 5, John Dalton St.GLASGOW: 157a, St. Vincent Street. LEEDS: Pearl Chambers, East Parade.NEWCASTLE-UPON-TYNE : Pilgrim Heuse, Pilgrim Street.


17/128

CONTENTS Continued.CHAPTER V

CASE-HARDENING 56IMPORTANCE OF CASE-HARDENING CARBURIZING COMPOUNDSSELECTING A CARBURIZING COMPOUND CARE REQUIRED IN PACK-ING PARTS TO BE CASE-HARDENED TlME AND TEMPERATUREFACTORS IN CARBURIZING OPERATIONS HEAT TREATMENT SELEC-TION OF MATERIAL.

CHAPTER VIEQUIPMENT USED IN THE HEAT TREATMENT OFMETALS - - 63

LAY-OUT OF HEAT TREATMENT PLANT FURNACES : MUFFLE ANDOVEN TYPES RICHMOND NATURAL DRAUGHT REGENERATIVE OVENFURNACE LOW PRESSURE GAS AND AIR FURNACE SALT BATHFURNACES OlL TEMPERING BATH LEAD BATH FURNACEQUENCHING TANKS AND QUENCHING MEDIA SUPPLY OFTOOLS.

CHAPTER VIICONTROL OF HEAT TREATMENT TEMPERATURES- 73

IMPORTANCE OF PYROMETER RECORDS THE FOSTER PORTABLEPYROMETER THE AUTOMATIC RECORDER THE RADIATION PYRO-METER DETERMINING THE THERMAL CRITICAL POINTS OF STEELTHE SENTINEL PYROMETER THE CALIBRATION OF PYROMETERS.

CHAPTER VIIITHE APPLICATION AND CLASSIFICATION OF AIR-CRAFT METALS - - 82

DETAILED CLASSIFICATION OF AIRCRAFT METALS CASE-HARDENINGSTEELS : O.IO PER CENT. CARBON STEEL, O.I5 PER CENT. CARBONSTEEL, O.25 PER CENT. CARBON STEEL, LOW CARBON 3 PER CENT.NICKEL STEEL, LOW CARBON 5 PER CENT. NICKEL STEEL, LOWCARBON NICKEL-CHROME STEEL ORDINARY PLAIN CARBON STEELS I0.2 PER CENT. CARBON, 0.35 PER CENT. CARBON, MILD SHEET STEEL-ALLOY STEELS : 3 P'ER CENT. NICKEL STEEL, 5 PER CENT. NICKELSTEEL, NICKEL-CHROME STEEL, 25 PER CENT. NICKEL STEEL, TUNG-STEN STEEL NON-RUST STEEL STEEL FOR TUBES TESTS FORTUBE STEELS CARBON STEEL FOR AXLE TUBES NlCKEL-CHROMESTEEL FOR AXLE TUBES COPPER TUBES BRASS TUBES ALUMI-NIUM TUBES DURALUMIN TUBES STREAM-LINE WIRES BEARINGMETALS WHITE METAL FOR BEARINGS PHOSPHOR-BRONZE CAST-INGS FOR BEARINGS IRON CASTINGS FOR VALVE GUIDES AND AIR-COOLED CYLINDERS IRON CASTINGS FOR PISTONS AND WATER-COOLED CYLINDERS CAST-IRON PISTON RINGS ALUMINIUMENAMELLING ALUMINIUM CYLINDERS DURALUMIN.

1 1


18/128

THE AMALGAMS Co, LTD.Analytical and Consulting Metallurgists

Attercliffe Road New Eldon ChambersSHEFFIELD Cherry Street, BIRMINGHAMTelephone : Central 3665 Central 2938Telegrams : " Amalgam Sheffield " " Amalgams Birmingham "

Undertake work in the following branches :Chemical AnalysisMetal-microscopy and MicrophotographyMelting and Freezing Point DeterminationsCritical and Hardening TemperaturesExperimental Heat TreatmentTensile Tests Impact Tests Ball TestsInvestigation of DefectsConsultations in all Branches of Metal-

lurgy

SOLE MAKERS of the " BREARLEY "SENTINEL PYROMETERS

RANGES from 220 to 340 C. and from 600 to 1070 C.In Boxes of 50 of each Melting PointLISTS ON APPLICATION

Hardening and Tempering Salts, etc,12


19/128

CONTENTS Continued.CHAPTER IX

AUTOGENOUS WELDINGREPAIRING WORN OR BROKEN STEEL PARTS RELIABILITY OFCONSTRUCTION AND WELDING AREA OF ALTERED MOLECULARSTRUCTURE AUTOGENOUS WELDING, REALLY A CASTING ALTER-NATING STRESSES IN AIRCRAFT COMPONENTS EACH WELDINGPROBLEM TO BE JUDGED ON ITS MERITS PRECAUTIONS IN WELDINGPREPARATION OF PARTS FOR WELDING CONTROL OF EXPANSIONAND CONTRACTION WELDING RODS AND WIRES FLUXESWELDING CAST-IRON WELDING STEEL HEAT TREATMENT OFWELDS WELDED jOINTS INVESTIGATION ALUMINIUM-ZINCALLOYSFLUX POWDERS FOR ALUMINIUM ALLOYS USE OF FLUX FORALUMINIUM ALLOYS CONSTITUTION OF THE OXY-ACETYLENE FLAMEPREPARATION FOR WELDING ALUMINIUM ALLOYS EXPANSION ANDCONTRACTION OF ALUMINIUM ALLOYS IMPURITIES IN ALUMINIUMALLOYS SOLDERS FOR ALUMINIUM A.W.P. PROCESS OF WELDING.

100

INDEX - in

PYROMETERS/orALL DEPARTMENTSOF METALLURGY

TESTINGMACHINESFOR METALSCATALOGUES POST FREE

FOSTERINSTRUMENT CLETCHWORTH. ENGLAND


20/128


21/128

METALSIn Aircraft Construction

CHAPTER IINTRODUCTION.

1. The Work Of the Metallurgist. A working knowledge ofthe practical and scientific manipulation of aircraft metals must,obviously, be invaluable to the aircraft designer and constructor.Such appreciation of the work of the metallurgist is necessary toboth, in order that they may be able to place their requirementsclearly before him, and, also, that they may be in a position tohandle, to the best possible advantage, the metal he produces.To help to bridge over the gap between the laboratory of themetallurgist and the workshop of the aircraft constructor is theobject of this book.The scientist in his laboratory is apt to develop the physicalproperties of steel from an almost purely theoretical point of view,often ignoring the practical difficulties which confront the indus-trialist in his workshop. On the other hand, many users ofsteel, because of their inability to utilize the scientific data alreadyavailable, do not profit, to the full extent, from the researches ofthe scientist.Although the steelmaker, working from well-known facts anda large accumulation of scientific data, can, with certainty, con-

trol the chemical composition of his material, the matter of heattreatment still remains obscure in many particulars. But enoughis already known about the influence of temperature and themethod of its application, both in the processes of manufactureand on the finished products, to show the importance of this aspectof the subject.

2. The Selecting and Handling of Metals. However wellprepared a steel may be, it still requires caution in handling.The practical engineer, indeed, is faced with many problems.


22/128

First he has to se7efr the best : economically available steel forhis immediate purpose, and then he has to study how to use it.A steel which may, apparently, be ideal for certain purposes ofmanufacture, may have to be set aside as unsuitable, either owingto difficulties in working, high cost, or insecurity as regardssupply. Its behaviour, under workshop manipulations, and, later,under variations of temperature and stresses in actual use, de-mands close consideration. There should, therefore, betweenair pilot, aircraft constructor and steel maker, be an intimateexchange of views and experience which would result in theincreased reliability and safety of all types of aircraft.

3. Special Steels: Ternary and Quarternary Steels. Thesubject of special steels is one of wide application. The term"

special steel" is used to denote a -steel which contains one or

more special elements, such as nickel, chromium or vanadium incombination with carbon and iron. Special steels are divided intotwo groups, viz., those known as ternary steels, which containbut one special element alloyed with the iron and carbon; andthose known as quarternary steels, which contain two specialelements in addition to the iron and carbon.Each of the special elements has a different effect upon theultimate physical properties of the steel. The special elementswhich are added to give increased strength to structural steelsare nickel, nickel-chromium and chromium-vanadium; and thesteels containing these elements are those which are chiefly usedin the manufacture of aircraft component parts. Considering thatthey possess widely different properties, it is an exceedinglydifficult problem to estimate which class of steel is the most suit-able for a given part.

4. Advantages of Special Steels, On account of theirgreater toughness, special steels are more advantageous thanordinary carbon steels for constructional purposes; also, specialsteels are more readily adaptable to varying physical conditions.Ordinary carbon steels can be made stronger by increasing thepercentage of the carbon content, or by heat treatment, but boththese processes render them too brittle for use in aircraft con-struction.The chief characteristic of special steels is the variety, andindeed contrariety, of their physical qualities, with their infiniteadaptation to almost any specific requirement of construction.They can be made strong and tough at the same time. Bymaintaining the same strength combined with the additionaltoughness, parts can be made considerably lighter in section.This is of great importance in aircraft construction, where weightrequires every consideration with regard to the speed, enduranceand general structural stability of the machine.

5. The Production of High-Tensile Alloy Steels for Air-craft. The development of aircraft steels is of increasingindustrial importance. To realize thoroughly the practical possi-bilities of these steels under service conditions, it should beremembered that their characteristic properties are produced and16


23/128

controlled by precise and skilled handling during manufacture.The present age is one of specialization, and, so far as the steeltrade is concerned, it is a well-known fact that the manufactureand treatment of high-tensile alloy steels for aircraft work nowforms an important branch of the industry. Great progress hasbeen made in the production of these steels, with the result thatthe aircraft designer is now able to produce machines possessingconsiderable strength with a minimum of weight.

6. Scientific Heat Treatment of Steels. In recent yearsgreat progress has been made in metallurgical science. This hasresulted in almost revolutionary changes taking place in the heattreatment of steel in the workshop. Formerly, before heat treatmentwas scientifically applied, the physical condition of a given steel wasnot taken into consideration, but its chemical composition wasregarded as the criterion in selecting it for a given specifiedpurpose. Recent advances, however, tend to show that althoughthe chemical composition influences, fundamentally, the proper-ties of all steels, yet it is the manner in which they have beenphysically treated that, ultimately, fit them for any special pur-pose. If a steel is chemically pure it is generally supposed to befitted for the prescribed requirements; but, no matter how pure asteel may be chemically, unless it receives proper treatment inthe subsequent working, it will be unable to stand the require-ments of practice. There are instances on record where a chemi-cally pure steel, employed for the same purpose, and used in thesame manner, as a steel containing a large percentage of impuri-ties, has, owing to improper treatment, failed under workingconditions to give as good results as the latter.Chemical impurities, such as sulphur, phosphorus and nitrogen,unless present to an abnormal extent, have very little effect uponthe physical properties of steel, provided that these impuritiesare distributed uniformly enough to maintain chemical homo-geneity. The presence of impurities is only injurious when theysegregate into " ghost lines." These "ghost lines" usuallycontain a large proportion of carbon, phosphorus and manganesesulphide, thereby producing a mechanical separation, causing thesteel to be weak and brittle.Although in some cases failure is due to steel of unsuitablechemical composition, in most cases it is traceable to impropertreatment. A slight difference in the chemical composition is

hardly noticeable, but a slight difference in the heat treatment willsoon be detected, particularly in the hardening and temperingprocesses.Needless to_say, the value of correct heat treatment is fullyrecognized in the aircraft industry. The new steels which havebeen introduced to meet the exacting demands of this section ofsteel users, can, under different treatments, be made to possessa wide range of mechanical properties. The same steel underdifferent treatments, may exhibit extreme hardness with verylittle ductility, or an intermediate hardness combined with a fairamount of ductility, or extreme ductility together with softness.These varying conditions are entirely brought about by a judiciousapplication of heat treatment.

17


24/128

CHAPTER IITESTING THE STRENGTH OF MATERIALS

7. Special Importance of Strength of Materials to Air-craft Designers. The subject of the strength and reliability ofthe materials used in aircraft construction has always been ofconsiderable importance to the designer, and this importance willcontinue to increase with the development of commercial aviation.The employment of the best suited material in the manufactureof the various parts which constitute an aircraft not only leads tothe increased reliability and longer life of the aircraft structure,but also tends actually to reduce the cost of aerial trans-portation.With regard to this latter aspect of the commercial importanceof employing the best obtainable material, much could be said, butit will, here, suffice to recall the fact that the more suited amaterial is for a given part, the lighter that part can be made,and, by so much, will the dead weight of a machine be reduced.And, if it be remembered that in an aeroplane there are hundredsof small parts, it will be easy to realize what a considerable weightwill be saved, in the aggregate, if on each part a reduction besecured by using the best material obtainable for the particularpart. This reduction in the dead weight of an aeroplane structureis immediately attended by the capability of the machine to carrya greater useful weight.

Again, by the use of the best available materials, it is possibleto diminish the thickness of a wing structure, or the dimensionsof a fuselage, or the size of the constituent parts of an under-carriage, or the magnitude of any other portion of the machinewhich offers resistance to penetration through the air. Such areduction would lead to an increased fineness of the machinewith all the resultant gain in the power required to drive it, andin the ease with which the various manoeuvres necessary for aerialnavigation can be carried out.

8. Stresses Induced in Aircraft Parts. All the materialsused in the construction of aircraft should be capable of with-standing certain definite loads. These loads, which are theexternal forces applied to the machine during flight, are termedstresses. Stress, therefore, may be defined as a force, or a com-18


25/128

bination of forces, which a body may have to withstand underworking conditions. The effect of a load, or stress, upon apiece of material is to change, or tend to change, the shape ofthe material. The change of shape is called deformation, and,as it will be seen later on" each kind of load has its appropriatedeformation.Most aircraft parts have to resist more than one kind of stress,that is to say, the parts are called upon to withstand workingloads applied in different planes with reference to the central axisof the object. The engine crankshaft may be taken as a typicalexample of a part subjected to combined stresses. When theengine is at work, the material of the crankshaft has to undergobending and twisting stresses while, at the same time, it has towithstand shock. The material of the crankshaft should, there-fore, be able to resist these various stresses and should, besides,possess the requisite amount of hardness to withstand unduewear.An exact knowledge of the practical working conditions of eachparticular part of the machine, in both normal and abnormalflight, is indispensable to the aircraft designer if he is to securethe best results possible from any type of aircraft. The magni-tude of the stresses, and the manner in which they are applied,should be carefully considered in the designing of each part, andthe material selected for use in the construction of the part shouldbe thoroughly capable of resisting the particular kind of stress towhich the part is subjected in practice.

9. Variation in the Properties of Materials, and the Import-ance Of Testing. The properties of the materials used byengineers vary through many causes. If we take a pound ofalmost any gas and measure its volume and the temperature andthe pressure to which it is subjected, we can estimate withaccuracy the behaviour of the gas if it is to be compressed, undercertain conditions, to a new volume. If, however, we take apound of steel, we cannot forecast with certainty the behaviourof the material under a load, if we know nothing more about thematerial than the fact that it is called steel. It is of some assistanceto us if we know the chemical composition of the steel, but, evenwith such information, we cannot estimate with accuracy the loadwhich will cause the material to fracture.

Experience has taught us that the only satisfactory methodof estimating suitable loads in the designing of machines is tosubmit to tests samples of the actual material to be used, whichtests reproduce, as nearly as it is possible, the conditions metwith in actual practice. Thus if a part is to be worked underconditions of direct pull, the most satisfactory way of decidingthe value of the material to be used, is to subject samples of it toa test reproducing these conditions. It is of great importancethat the specimens selected for testing purposes should bethoroughly representative of the material under inspection, andthat they should undergo exactly the same heat treatment pro-cess as the parts which are eventually to be made from thematerial represented.

19


26/128

10. The Various Tests applied to Aircraft Steels. Owingto the necessity of obtaining comparable results, it is importantthat standard methods of testing the materials should be adopted,as the test values are affected considerably by a difference in thesize or shape of the specimen, and the manner of testing. Thetests applied to aircraft steels may be classified under the follow-ing headings :

(1) HARDNESS.(2) TENSION.(3) IMPACT.(4) BENDING.(5) FATIGUE.

It is absolutely essential that one or more of the foregoingtests should be applied to each consignment of material beforecommencing to manufacture the parts, but, once the suitability ofthe material has been established, it is not necessary to apply allthe methods of testing to each batch of parts. Thoroughlyreliable information can be obtained from the use of simple andinexpensive methods. To test the final properties of the finishedparts it is only necessary to check a proportion of them by theBrinell hardness method.The relation between the hardness and the other properties ofsteels does not vary if they are of the same composition, andreceive the same treatment. Therefore, provided that a hard-ness number, identical with that obtained in the preliminary testspecimen is obtained on the finished part, it can be safely assumedthat the final properties of the part are identical to those of thetest specimen.

11. Hardness Testing. The relation of hardness to otherphysical properties calls for a careful and detailed study. Hard-ness and ductility are the most important properties of high-tensile steels. In these steels high hardness combined with highductility together form an alloy of great strength, introducing intothe material a further important physical condition characterizedas toughness. To secure the maximum efficiency in the finishedparts, it is essential that hardness and ductility should be combinedtogether in proper ratio. The hardness of a solid substance maybe defined as the resistance offered by a smooth surface of thesubstance to abrasion or deformation. Hardness measured bythe file denotes in a simple, yet practical, way the power of theobject to resist the wearing away of its surface under the cuttingaction of the file teeth. In many cases the practical man, for wantof more scientific means, has to resort to the file in order todetermine whether, or not, the object is of the requisite hardness.It must be admitted that, although in many instances the intelli-gent practitioner obtains wonderfully successful results from itsuse, the limitations of the file are obvious; too much depends onthe nature of the file, and the method of application is alwaysliable to variation.12. The Brinell Ball Machine and Hardness Number.

Hardness, as measured by the various standard hardness testinginstruments, is based on the resistance of the material to pene-20


27/128

tration and deformation. Of the instruments in use the Brinellball machine is perhaps the most satisfactory for general purposes.It has a universal application; and, by engineers and steel makers,it is arbitrarily accepted as a standard instrument. Hardnesstests carried out by the ball machine are known as static hard-ness tests. A Brinell ball machine is shown in Fig. i. The testsconsist in pressing a 10 m/m diameter hardened steel ball undera known pressure into the surface of the material, and measuringthe area of the impression. The area of the indenture variesaccording to. the resistance to deformation of the material beingtested. The result of the test is expressed in hardness numeralswhich are calculated from the following formula : .

TOTAL PRESSURE.HARDNESS NUMBER = CURVED AREA OF IMPRESSION.For practical working,machines are designed to

give a standard pressure of3,000 kilos. The diameterof the impression is mea-sured by a micrometer mi-croscope, and, by means ofthe diameter indicated, thehardness number may beread oif at a glance from ascale supplied with the ma-chine. It must be noted thatthe accuracy of the test de-pends entirely on the correctmeasurement of the diameterof the impression. In aworks machine, the hardenedball, owing to its frequentuse, is liable to bear a num-ber of flattened surfaces onthe periphery.Unless the ball is fre-quently replaced, the impres-sion in the samples will notbe a perfect circle, thuscausing liability "of errorin the subsequent measure-ment.Table I (p. 22\ shows the

Brinell diameters of impres-sion with correspondinghardness numbers, and alsothe approximate maximumtensile strength.

FIG. i.BRINELL BALL HARDXESS

TESTING MACHINE.21


28/128

TABLE I OF BRINELL DIAMETERS OF IMPRESSION,WITH CORRESPONDING HARDNESS NUMBERS AND APPROXIMATE MAXIMUM

TENSILE STRENGTH, USING A 10 M/M BALL.

3,000 Kg. Load.


29/128

A high-class microscope and a Table of Hardness Numbers areincluded with each machine.

FIG. 2.THE AVERY BRINELL HARDNESS

TESTING MACHINE.

14. The Shore SclerOSCOpe. Another form of hardness testsometimes adopted is the one carried out by means of the ShoreScleroscope. This does not replace the Brinell machine, but it isvery often used as a supplement to it. The apparatus, which isshown in use in Fig. 3, consists of a vertical graduated glasstube, which contains a small diamond-pointed drop hammer. Bymeans of a suction bulb a vacuum is produced in the glass tube,and the hammer is drawn up to the top, where it is held inposition by hooks. By applying pressure to the bulb, air iscaused to act on the valve operating the suspension hooks, andthus the drop hammer is released. Th>e height of the fall is teninches. The sample to be tested is solidly mounted under thehammer, which on being released strikes the surface of the test-piece and quickly rebounds up the graduated tube. The positionof the top part of the hammer should be noted on the scale whenit comes to rest, and before it again commences to descend.23


30/128

It is rather a difficult matter at first to get reliable readings,but after some experience fairly consistent results can be obtained.The surface of the test-piece should be properly prepared in order

FIG. 3.SCLEROSCOPE IN USE, TESTING GEARS.

to secure reliable results. It is absolutely imperative that thedecarbonized surface should be removed before testing, otherwisethe reading will be much too low. Extremely hard surfaces mustbe free from tool marks, and should be perfectly smooth, but mustnot be highly polished. It is not essential, however, to so care-fully prepare medium-hard and soft surfaces. Generally speaking,finishing off the surface with a second cut file is sufficient.The surface of the sample on the point of contact with thehammer is considerably strained, and the hammer- should not,therefore, be allowed to drop more than once on the same place,as the reading will be then too high. It must not be assumed thatthe strained point has any detrimental after effect on the efficiencyof the object, for the affected part is so minute that it has verylittle, if any, influence on the strength of the object tested.15 B Tensile Tests. The tensile test shows the following items

of information : (i) the elastic limit, (2) the yield point, (3) themaximum stress, (4) the extensibility, (5) the reduction ofarea.

In practical works testing, the elastic limit and yield point areexpressed as one item. Here, however, they will be considered-separately because, to some extent, they are two distinct pointsof the testing operation. Fig. 4 shows a general view of theBuckton 5o-ton Vertical Testing Machine.When the load is applied to the specimen the piece increasesin length proportionally to the force applied, until a point isreached at which the increase in length ceases to be proportionalto the loading. At this point, if the load is removed, the specimenwill return to its original dimensions; the specimen is in no wise24


31/128


32/128

permanently deformed. Up to this point stress and deformationare in constant ratio. Repeated stresses up to this point can beapplied, but each time the load is removed the test specimen willreturn to its former size. The maximum amount of force thatcan be applied without permanently altering the shape of the test-piece is known as the " elastic limit " of the material, and isexpressed in tons per square inch. If the applied stress exceedsthe elastic limit, the test-piece gradually decreases in the sizeof its cross-section, and at the same time increases in length.Once the elastic limit is exceeded, the material takes what isknown as a " permanent set." It no longer returns to the originaldimensions when the load is removed.During the testing operation, if the distance between the gauge

points on the specimen is carefully measured by dividers, theactual point, at which the increase in length becomes visible, isknown as the ff yield point." The load on the specimen is knownas the yield stress, and is expressed in tons per square inch. Thepoint can- also be distinctly observed by the sudden drop of thebeam of the testing machine. As the load on the test-piece in-creases, the weight is moved along the beam to just balance it.When the yield point is reached there is a sudden elongation ofthe test specimen, and the load on the machine is arrested to suchan extent that it causes the beam to drop. It only remainsstationary in this position for a few seconds, but is easily observ-able. It lasts until the movement of the pulling force catches upto the elastic flow of the specimen.The load should be uniformly applied, without intermission,until the test-piece actually breaks; it should never be taken offand re-applied, but should always be maintained continuously.The accurate determination of the yield point is of considerableimportance. The yield stress is the maximum safe working loadfor all parts working in tension. Therefore the suitability of amaterial for any part that is to be worked in a state of tensionis dependent on the yield point. Although a part may be stressedbeyond the yield point without actually breaking, at this stage thematerial is more or less in a plastic condition. Consequently, ifthe working stresses exceed the yield point, there is a greatliability of the part becoming permanently deformed. On furtherincreasing the load on the test-piece, the specimen continues tostretch.The test-piece stretches uniformly, along the parallel portion,

until the load nearly approaches the breaking point. Just beforebreaking there is a decided decrease in the section of the pieceat one particular point. When well developed, this point has anappearance very much like that of a bottle neck. Fig. 5 shows atypical example of this " necking." Almost immediately after the" necking " is observed, the specimen fails to lift the beam : themaximum stress has been reached. This is expressed in tons persquare inch. The actual increase in the length of the test-pieceis known as the elongation of the material, and is expressed asso much per cent on the original length of the parallel portion.The percentage of the elongation denotes the ductility of thematerial.


33/128

When expressing results, it is important that the distance mea-sured between the gauge points on the parallel portion shouldalways be stated, because, even for the same material, the per-centage of elongation will be larger on a short specimen thanit would be on one having wider gauge points. This is explainedby the fact that there are two distinct elongations : one of whichis uniform over the greater part of the parallel portion, and theother is confined to a shortlength in the region of the" necking." This latter partis only a comparatively smallportion of the actual lengthunder observation, and doesnot vary much in specimensof the same material, nomatter what the overall dis-tance between the gaugepoints may be. The amountof stretch in the region of the"necking " is much greaterthan that on the other partof the specimen. Assumingthat we have two specimensof the same material, onehaving a one-inch actinglength (as the distance be-tween the gauge points iscalled) and the other a two-inch acting length, we shouldhave the same amount ofelongation in the region ofthe neck on each specimen.Therefore, as the actualstretch in this part is muchgreater than that on the re-mainder of the parallel por-tion, it follows that therewould be a higher elongationper cent on the shorterspecimen.

Just as specimens havingdifferent acting lengths givedifferent elongations, so speci-mens having the same actinglength will give the sameelongation only if they are ofthe same diameter. Thusa test-piece 0.5 inch diameter by 2 inches acting length wouldnot give the same elongation as a test-piece 0.564 inch diameterby 2 inches acting length. It will be seen that it is highly im-portant that the tensile specimen should be of standard dimen-sions, i.e., the diameter and acting length should always be pro-portional to each other. 27

FIG. 5.TENSILE TEST SPECIMENS, BEFORE

AND AFTER TESTING.


34/128

Fig. 6 shows a very useful and compact form of testing machinemade by Avery, Birmingham. This machine is fitted with avariable speed motor, and has an autographic recorder. It canbe adapted for testing specimens in tension, compression, bending,shearing, torsion and hardness.

16. Standard Form of Test- Piece. The standard form oftest-piece adopted for high-tensile steels is 0.564 inch diameter by2 inches acting length. When, unavoidably, the material is

FIG. 6.AVERY 3O-TON TESTING MACHINE.

too small to allow of this standard size of specimen being pre-pared, the smaller dimensions used should be in the same pro-portion. These may be calculated from the formula :

ACTING LENGTH/ = = 4V AREA OF SECTION

Figs. 7, 8, 9, 10 and n show the various forms of tensile testspecimens in general use.Whatever size of specimen is adopted it should be preparedwith the greatest care and accuracy. In the case of a hardmaterial, such as for instance a loo-ton steel, the angle betweenthe parallel and the shoulder should be nicely radiused, otherwisedifficulty will be experienced in the manner in which the specimenbreaks. If the vertex of the angle is sharp, the specimen isnearly sure to break in the sharp corner.8


35/128

FIG. 7.FORM OF TENSILE TEST SPECIMEN.

(Note. When the ends are i inch or i^ inch diameter,the shoulder shown .625 inch diameter may be increased indiameter if desired.)

The amount of contraction in the section of the test-piece at thepoint where the fracture takes place is expressed as a percentageof the area of the original section, and is specified in the test resultas the reduction of area. Together with the elongation, the reduc-tion of area indicates the ductility of the material. Moreover, itusually follows that if a material has a good reduction of area,it will also give a good impact result.


36/128

4-24'Dta

*-FIG. 9.FORM OF TENSILE TEST SPECIMEN.

(Note. The ends are to be turned in order to ensure thatthey are concentric with the specimen, but " D " should bekept as large as possible.)

17. Impact Tests- The importance of the impact shocktest has, of recent years, come into great prominence in Govern-ment specifications, largely on account of the exacting demandsmade by war conditions on the science of metallurgy. As would be&t JP

i / *


37/128

FIG. 12.IZOD IMPACT TESTER.

The two forms of impact test most generally used are the Izodand the Charpy. Both methods employ a small notched speci-men which is tested to destruction. The Izod machine shownin Fig. 12 is the one generally employed in Great Britain. Repre-sentative samples of the bulk of the metal parts used by aircraftand automobile manufacturers, working to Government specifi-cations, are required to comply with pre-determined standards ofIzod value before the parts are assembled.

18. The Izod Machine. The Izod machine consists of a heavypendulum swinging in a vertical plane about the point of sus-pension. The pendulum swings on ball bearings, and developsan energy of 120 ft.-lb. The vice fixed on the bed of the machinec 31


38/128

holds the test specimen in the manner shown in Fig. 13. Thelocation is such that the specimen is held directly in the pathof the pendulum. A graduated arc is provided over which movesthe pointer. The pointer is rigidly fixed to the pendulum, thusgiving an indication of the angular position of the pendulum atany point of its swing. In making a test with this machine thependulum is swung up to a definite pre-determined angle, and heldthere by a clamp. The specimen is fixed in the vice with thenotched side towards the pendulum. When the pendulum isreleased, it swings down; and when it is at the lowest point of

FIG. 13.VICE OF IMPACT TESTER, SHOWING SPECIMEN

IN POSITION FOR TESTING.its fall, it strikes against the specimen and breaks it. It thencontinues on its swing with its momentum more or less diminishedaccording to the amount of energy required to break the speci-men. By virtue of the law of conservation of energy, if there hadbeen no specimen to obstruct it, the pendulum would, neglectingslight frictional effects, have risen on the other side to a heightequal to that from which it was let fall. The specimen in beingbroken absorbs a certain portion of the pendulum's energy, andthis is measured by the continued and diminished swing of thependulum moving the pointer over the graduated scale. Thegreater the resistance offered by the specimen, the less the pen-dulum will continue in its swing, and so the pointer will becarried over the graduated quadrant a shorter distance.

Fig. 14 shows the form of test specimen which is machined tostandard dimensions. It is provided with a notch. The presenceof the notch concentrates the stress at one point relative to thepoint of contact with the pendulum; otherwise extremely soft orductile materials would bend over. Specimens are usually madethree on one bar, the bar being raised in the vice after eachone has been broken. It is essential, in erecting this machine,to bolt it down on to a concrete base. If this is not done, thetest results will be affected, and will not be strictly comparablewith tests made on other Izod machines in use.

3 2


39/128

19. Value Of Impact Test. The reasons which have broughtabout the use of the impact test x and its usefulness in connectionwith the construction of aircraft, will now be considered briefly.The impact test is applied because it is the test which gives mostinformation as to whether the heat treatment, which has beengiven to the steel during the manufacturing processes, is satis-factory or not. If the material under consideration has beenheat treated to the best advantage, the Izod value will be rela-tively high. If, on the other hand, the heat treatment given hasbeen inferior, then the Izod value will be relatively low.

- Flat 'A'Hte

*-ENLARGED VIEW OfMOTCH

1-005 radius

The impact results have no relation to either the tensile or theBrinell hardness tests. The fact that a material is said to givean impact result of 50 ft.-lb. is no criterion that the material inquestion is either good or bad. Other physical properties and thechemical composition must accompany the test before the useful-ness of the material can be judged. If in an alloy steel the impactof 50 ft.-lb. were accompanied with a tensile strength of 75 tonsand a ball hardness of 350, then the material would be extremelygood ; if, in the same class of material, it was only associated witha tensile strength of 40 tons and a ball hardness of 170, then thematerial would be decidedly poor. On the other hand, if animpact of 50 ft.-lb. along with 40 tons tensile and a ball hardnessof 170 were obtained from an ordinary plain carbon steel con-taining, say, from 30 to 40 per cent carbon, then the impact figurewould be a remarkably good one." Brittleness," in machine construction, is a feature, whichmust be well considered; otherwise it is likely to become dan-gerous and cause failure at the critical moment. For instance,the study of a large number of failures of engine parts goes toshowr that steel with a bad Izod value fails more often than steelwith a good Izod value. Brittleness may be due to the steel con-taining too high a percentage of sulphur and phosphorus. Thisis a very common occurrence in steels used in making nuts andbolts. If the sulphur and phosphorus content is over i per cent,


40/128

the Izod value obtained would be in the region of 2 to 3 ft.-lb.,whereas a similar steel with sulphur and phosphorus below .06 percent would give an Izod test figure of over 40 ft.-lb.20. Heat Treatment and Izod Result. An interesting demon-stration of the effect of heat treatment may be gathered from theIzod result. The following tests taken on a 5 per cent nickel

steel clearly show the remarkable increase in toughness broughtabout by judicious treatment :

5 % Nickel SteelSpecimen.


41/128

made in capacities of from 25 to 200 kilogramme-metres, thelarger sizes being in some cases furnished with electric control.Fig. 1 6 gives a diagrammatic representation of the form ofthe Charpy test specimen.

FIG. 16.DIAGRAMMATIC VIEW OF CHARPY IMPACT

TEST SPECIMEN.22. Impact Fatigue Tests: The Stanton and Eden-FosterMachines* The machines used for carrying out impact fatigue

tests are the Stanton and the Eden-Foster. Both machines are verysimilar in application. In principle, they attempt to combinethe effects of the vibration of pure fatigue test and the impact orshock test, by testing a specimen to destruction by a method whichproduces both shock and reversals of stress. The results of testsby the Eden-Foster method exhibit a consistency not hithertofound in either the vibration or the shock test ; therefore it appearsthat this test supplies something previously lacking in both thesetwo methods, when carried out separately.The Eden-Foster test, in its simplest form, consists in repeatedlydropping a hammer of known mass from a given height on to aspecimen supported at either end in a horizontal position directlybelow it, the specimen being further arranged to rotate about itsaxis through 180 degrees between each blow.The essential feature in which this method differs from thevibration method is that the stress is not applied gradually, butinstantaneously, thus obtaining the " shock " effect. On the otherhand, the method is something more than a mere impact test,since the rotation of the specimen between each blow produces, aswill be explained later, a reversal of stress throughout the cross-section of the specimen.

Figs. 17, 18 and 19 show the Eden-Foster machine. The wholeof the mechanism is secured to, and supported by, the main cast-ing, which also acts as cover to the tank or box casting. Pro-jecting through the side of the tank, but not seen .in the illustra-tion, is the main spindle, which may be driven by a one-inch beltfrom any convenient countershaft, or alternatively, by an electricmotor with suitable gear or worm reduction. About one-tenthof a horse power is required.

35


42/128

The main spindlecarries a dog clutchnormally in engage-ment and driving acam. A roller bearson the upper surfaceof this cam, and isattached to thepower end of the rodH, being suitablyguided so that itrises and descendsat each revolution ofthe cam. Fixed onthe rod H is an arm] , which engageswith the lower faceof the hammer M,so that when the rodH rises by rotationof the cam the ham-mer M is also lifted.The hammer slidesfreely between twosets of three-pointguiding screws,these screwsbeing covered by thetwo castings which,as shown in the illus-trations, are at-tached to the stan-dard G and its fel-low on the oppositeside.Mounted upon thestandard G is a sleeveW, free to rotate

FIG. 17.EDEN-FOSTER IMPACT TESTINGMACHINE.

about the standard, but normally held in a fixed position by thespring L. Clamped on the sleeve W is an adjustable catch K. Assoon as the arm J has lifted the hammer M sufficiently, the spring Lcauses a partial rotation of the sleeve W, so that when the arm Jagain descends the hammer is held by the catch K. The furtherdescent of the arm J brings its lower inclined face in engagementwith roller arm N, attached to the sleeve W, in such a mannerthat the catch K releases the hammer M, allowing it to fall uponthe test-piece "O, which is carried by two hardened steel bushes inthe plummer blocks P.As already stated, the test-piece is rotated through 180 degreesbetween successive blows, this being effected as follows :One end of the test-piece is slotted to engage with a universaljoint drive, and is kept in engagement by a screw R. The univer-sal joint is driven through a free-wheel and clutch T, by a length

36


43/128

of chain S, one end of which is attached to the roller alreadyreferred to as bearing on the cam, the other end carrying a weightsuitably guided. As the roller moves upwards the weight on theother end of the chain draws the chain over its sprocket, thusrotating the test-piece. When the hammer sinks, the chain andsprocket, of course, return to their original positions, but the free-

FIG. 18.EDEN-FOSTER IMPACT TESTING MACHINE.

37


44/128

T

FIG. 19.EDEN-FOSTER IMPACT TESTING MACHINE.

wheel " slips " on this backward motion, so that the test-pieceis not turned back with them.

It will be observed also that the rotation of the test-piece beginsand ends entirely between successive blows. The revolutions of38


45/128

the test-piece are recorded by a counter V, and, of course, thenumber of blows is found by multiplying- the counter record bytwo. When the test-piece breaks, it comes into contact with anarm X, and thereby trips the clutch on the driving shaft, andstops the machine. The tank is partially filled with oil to provideefficient lubrication of the cam and other surfaces.The hammer M is shod with a tip of hardened tool steel. Theheight of the drop depends on the position of the adjustable clutchK on the sleeve W, and may be varied from about i to 4! inches(25 to 113 m/m). To allow a wide range of tests, two hammersare provided weighing 5 and 2 Ib. (2.26 and 0.91 kilos) respec-tively. Using the 5>lb. hammer, the main spindle may be drivenat any speed up to about 60 revolutions, or with the 2-lb. hammerup to about 90 revolutions per minute, giving 60 to 90 blows perminute respectively.

FIG. 20.RESULTS OF BEND TESTS.

23. Bending Tests. Fig. 20 shows three bend tests, the frontspecimen broke at an angle of 90 degrees, the specimen on the leftbent over a radius equal to its own diameter to an angle of 180degrees without showing signs of fracture. This is the test usuallyexpected of aircraft material with a tensile stress up to sixty tonsper square inch. In commercial work the test is discontinued atthis point; but, for experimental purposes, it is usually continueduntil the specimen either fractures or bends double as shown onthe right.

39


46/128

CHAPTER IIIDEFECTS IN STEEL

24. Basis for Proper Selection of Steel. The aircraft steelmaker is called upon to produce steels which must satisfy theconditions of very stringent specifications. In some cases theconditions are comparatively easy to attain, whilst in others someof the tests required are very difficult to accomplish, the materialrequiring1 expert manipulation during the manufacture. Consi-dering that these steels possess widely different mechanical pro-perties, it is a very difficult problem to decide which classof steel is the most suitable for a given part. It is possible, how-ever, to learn a great deal of their relative properties by makinga series of practical heat treatment and simple mechanical tests.The knowledge, gained by conducting such a series of experi-ments upon some of the various steels, provides a basis for theproper selection of the material best adapted for a given purpose.Before attempting experimental investigation, it is well toconsider the preliminary methods that should be recognized. First,a standard series of heat treatment temperatures should beadopted for each class of steel containing the same special element.By applying the same treatment it is possible to ascertain theeffect of varying amounts of the special element. The process ofheating and cooling the samples should be performed under condi-tions similar to those occurring in actual practice.

25. Workshop Difficulties. The field of investigation of themanufacture and treatment of special steels is by no means ex-hausted. Although steels of dissimilar chemical composition areable, by judicious treatment, to attain the same standard ofefficiency in the finished parts, some steels are liable to a largerpercentage of waste during manufacture. And, in order tominimize the risk of the material being faulty, it is necessary tobestow greater care in the forging, annealing and heat treatmentwhich consequently increases the cost. This increase in the costof production does not materially benefit the steel user. Fromthe aircraft constructor's point of view, the best steel to use, fora given purpose and high standard of efficiency, is the one thatcan be most easily obtained and replaced, and is easy to work and

40


47/128

machine. Experiments, pursued on practical lines, often advan-tageously show that a less expensive steel will prove quite assuitable for a specific purpose as one of higher cost that waspreviously adopted. It is clear, however, that economy in the firstcost of material is not to be recommended at the expense ofefficiency.

Unfortunately, the present tendency of most investigators is tofurther develop the physical properties of special steels from apurely theoretical standpoint, and to leave the practical ground-work and shop difficulties unsolved. Although it is admittedthat theoretical researches and developments are essential, yet,on the other hand, it is a matter for regret that more work has notbeen done with a view to finding new methods to help toalleviate the difficulties of thesteel user. Many of these work-shop difficulties have a simpleorigin, and, if understood, canbe easily remedied in the initialstages.

It is proposed briefly to con-sider these difficulties, and seewhich of them are due tooriginal defects in the rawmaterial. For this purpose itis necessary to have a goodknowledge of the defects thatmay be possible in steel bars,as delivered from the steelmaker to the aircraft manu-facturer ; otherwise operationsmay be commenced and a lotof needless and expensivemachining work done before itis realized that work is beingspent on faulty material.

FIG. 21.PIPE " IN NICKEL STEEL

INGOT.

26. Elimination of Sulphur and Phosphorus. The steelmaker is now thoroughly able to control the chemical composi-tion of his material, so that cases of failure due to the injuriouseffects of chemical impurities are now rarely met with. Sulphurand phosphorus are fully recognized by the steel maker as harm-ful ingredients, which are introduced into the steel from theraw material. Therefore special attention is usually paid in theselection of the basis material with a view to keeping the amountof these elements as low as possible. They are usually present insuch small quantities that their influence on the physical propertiesmay be almost entirely disregarded.27. Occluded Cases. During the process of manufacture, theintroduction of some gases, which exert an injurious effect on the

steel, is not very easily stopped. All steels contain gases held insolution, which are known as " occluded gases." Some of thesegases, the chief of which is nitrogen, exert a far more harmful

4*


48/128

effect than is generally recognized. The action of nitrogen,if present in a high percentage, is extremely injurious. Thematerial then possesses characteristic properties of general un-soundness. It cracks and breaksup during hot working. Evenif the nitrogen is present in onlysmall quantities in a steel, thematerial is liable to suffer inconsequence, and exhibits amarked tendency towardsbrittleness. The exact influenceof nitrogen on the physical pro-perties is not yet thoroughlyestablished, but it is generallyconsidered that the tensilestrength increases with the pro-portion of nitrogen, whilst theductility falls. If the percentagereaches 0.037 the ductility almostentirely disappears. On an aver-age the material will only giveabout 0.5 per cent, elongation.If, however, such a percentage is met with in actual practice, it isnot altogether safe to conclude that the material is valueless,because to some extent the deleterious influence depends upon theform in which the nitrogen exists in the steel. Some typical casesmay be cited where a steel containing 0.025 Per cent, of nitrogenhas given fairly good results in elongation. It is important, when

PIPEFIG. 22.IN NICKEL STEELINGOT.

FIG. 23." PIPE " IN JOURNAL OF AERO CRANK-SHAFT.the nitrogen content reaches the critical limit, carefully to testfor brittleness before use is actually made of the material.In the case of hardened steels the deleterious effect of thenitrogen is considerably increased. Special steels are particularlyliable to its injurious influence, and should, consequently, befreed from the element in the melting furnace.

Ferro-silicon is the chief cause of the occurrence of nitrogen in42


49/128

steel. During the melting process the additions of this compoundto the molten metal should be made with extreme care. At steelmelting temperatures, the nitrogen present in the ferro-siliconreacts on the silicon, and, unless the additions are carefully made,the nitrogen enters into solution with the iron. The" use ofvanadium is the safest means of eliminating nitrogen. At hightemperatures the vanadium combines with the nitrogen, and thusfrees the metal from the influence of this deleterious gas. Steelmade by the crucible process is far superior to that made by othermethods, for the simple reason that it contains very little occludednitrogen.

28. Elimination of Defectsin Steel Ingots: Piping, Blow-holes, RoakS, Laps. Steel in-gots always contain certain de-fects, which must be elimin-ated during the subsequentmanufacturing processes. Allthese defects are under the con-trol of the steel maker. By skilledmanipulation, during the work-ing-up processes, they can be en-tirely eliminated, so that, finally,it is possible to produce athoroughly homogeneousmaterial. The steel user shouldunderstand these preliminarydefects in order that he mayrecognize them in a material, inwhich they may be present,before he commences to use it.It sometimes happens that they unavoidably escape detection duringforging and rolling, and that, consequently, during the final stagesof manufacture of the steel parts considerable annoyance is caused.

Although modern methods cfsteel manufacture have reducedthe percentage of waste due topiping in the ingot to aminimum, all ingots, as cast,contain a certain amount ofpipe. It is the usual prac-tice of the steel maker todiscard a certain percentageof material from the top end ofthe ingot in order to ensure thatthe metal is free from this de-fect. The pipe is caused byshrinkage during the solidifica-tion of the molten metal. Figs.21 to 25 illustrate the appear-ance of pipe in steel. Fig. 21shows pipingin a 14-inch 5 percent, nickel steel ingot. ^Theappearance of its fractured sur- 43

FIG. 24.MlCROPHOTOGRAPH OF " PlPE "

IN TENSILE TEST-PIECE.

FIG. 25.MlCROPHOTOGRAPH OF " PlPE "

IN TENSILE TEST-PIECE.


50/128

FIG. 26.FRACTURE OF STEEL BAR SHOWING

SEGREGATED CENTRE.

face shows that the ingot was cast at the correct temperature. Fig.22 shows the fractured surface of a similar ingot; but, in thiscase, the temperature of cast-ing was much too high. Itwill be noticed that, in additionto the piping, bloiv-holes arepresent to a fairly considerableextent, and, moreover, the growthof crystallization is very apparentfrom the outer edge. Fig. 23shows the appearance of pipingin an aero-engine crankshaftjournal. This shaft failed in ser-vice. Figs. 24 and 25 are trans-verse and longitudinal sectionsrespectively through the pipefound in a tensile test-piece ofnickel-chrome steel.

Immediately below the pipedarea there is always a certainamount of segregation which isquite as objectionable as the pipeitself. The fracture of a steel bar which exhibits a hard centreis a characteristic example, showing the result of rolling down aningot containing an appreciable amountof segregation in the region immedi-ately below the pipe. One of the con-tributory causes of segregation is thecharcoal used for keeping the top ofthe ingot hot. It settles into theingot and highly carbonizes thetop portion. It is important that thetop of the ingot should be cut off wellbelow this unsound area. The re-maining portion will then, practicallyspeaking, be chemically homogeneous.Fig. 26 shows the appearance of afractured steel bar with a segregatedcentre.In present-day practice, blow-holesare not met with to any considerableextent in steel parts, but they are some-times present in cast iron parts, suchas cast iron cylinders or piston rings.In the case of small castings it is avery difficult matter to get rid of blow-holes, but in the manufacture of air-craft steels they are easily remediedduring manufacture by thoroughlycogging and welding the ingots. Fig.27 shows the appearance of blow-holesin the top half of a nickel steel ingot.To close up effectively these blow-holes,the honeycombed ingot should be44

FIG. 27.BLOW-HOLES IN NICKKLSTEEL INGOT.


51/128

FIG. 28.MlCROPHOTOGRAPH OF ROAK ONROUND STEEL BAR.

heated to a wash welding temperature say about 1,100 degreesC. and well soaked until a uniform heat is attained. Theheat should be sufficient to melt the scale on the surface. Aflux composed of sal ammoniac and borax is generally used.This helps to clean off the surface scale, and prevents the oxidationof the internal walls of the blow-holes. It is important to keep areducing atmosphere in the furnace during the heating operation,otherwise the walls of the cavities will be oxidized, and will conse-quently fail to weld up.In the subsequent rolling ofthe cogged bars the oxidizedsubterraneous blow -holessimply extend with the rolling,and appear on the surface of therolled bars as roaks, and pene-trate the bars radially to thedepth of the drawn out blow-hole. To some extent these de-fects can be overcome by chip-ping the surface of the coggedbars. When chipping is done,the bar should be nicelyshallowed out, and not, as issometimes the case, be cut toodeeply. The evil caused by deepchipping is quite as bad as theoriginal defect of the bar.The appearance of roaks on the surface of the rolled bars isone of the causes of cracking during the hardening operation.If the surface of the rolled bar is machined off before hardening,the presence of roaks may bepractically ignored, as the re-moval of the skin removes thedefect. However, in the case ofaircraft steels there always re-mains the liability of cracks,because most of the steels aretreated in the bar withoutmachining off the outer surface.Fig. 28 shows a microphoto-graph of a roak in a round steelbar.Laps are another form of sur-

face defect that may cause troublein the hardening operation. Theyoriginate in the rolling mill, andare due to the turning over androlling in of fash. Round barsare more liable to this defectthan those of any other section.The particular part of the rolling operation where the fault islikely to occur is when the bar is passed from the diamond tothe oval shape. On close examination it will be seen that lapsdiffer in appearance from roaks. They run obliquely into the45

FIG. 29.MlCROPHOTOGRAPH SHOWINGLAP IN ROUND STEEL BAR.


52/128

bar, and are always laid over. If examined under a powerfulhand lens they will be seen to enclose a layer of scale in theenvelope, formed by the laid over portion. .Fig. 29 shows theappearance of a lap on the top edge of a round bar of nickel-chrome steel.The presence of any of the foregoing defects renders the material

liable to failure in the hands of the steel user. Therefore it isadvisable for the user to examine thoroughly his material for thepresence of any of these original causes of unsoundness before hecommences to use it.


53/128

CHAPTER IV

THERMAL AND MECHANICAL TREATMENT OF STEEL29. Heating Processes. The steel, having been cast into

ingots, is forged in billets of suitable dimensions. The billets areeither drop stamped, or they may be rolled into bars of suitablesection for machining purposes. After drop forging or rolling, itis necessary that the material should be either normalized orannealed. The various parts are then machined, and afterwardshardened and subsequently tempered, to suit the particular re-quirements of the duties thoy may have to perform. Therefore,after the steel has been cast into ingots, there are four or fiveheating processes to consider before the material of which thepart is made is in a suitable physical condition to be assembledinto the machine.These processes are as follows :

IST. HEATING FOR FORGING.2ND. ,, ,, NORMALIZING.3RD. ,, ,, ANNEALING.4TH. ,, ,, HARDENING.5TH. ,, ,, TEMPERING.

30. Forging Operations. The close relationship existing be-tween the final physical properties of a steel and its thermal andmechanical treatment, should be clearly understood. Forging orrolling is not only necessary for working the material into therequired shape, but at the same time these processes change thestructure of the material into one of more uniform grain. A com-parison of Figs. 30 and 31 shows the refinement in the structureof a 0.30 carbon steel due to the closing up of the crystals byforging. Fig. 30 illustrates the microstructure, under a magnifica-tion of 200 diams., of a section cut from a 7-inch square forging,whilst Fig. 31 represents the microstructure after reducing aportion of this same forging down to 7~8th of an inch square.The physical tests corresponding to these micro-photographs

47


54/128


55/128

at all. After this period of lag has elapsed, the temperature of thesteel begins to rise again regularly. The procedure may be clearlyfollowed by means of the Diagram. Take the line A, C, M, thelag is indicated by the kink at C. When the metal is cooled slowlythe reverse phenomenon takes place. The cooling proceeds atfirst normally, until a point is reached at a temperature of about690 degrees C. Then, again, there is a period of seconds duringwhich the steel does not cool at all; it even may, during thisperiod, ge4: hotter, although all the time the furnace is actuallycooling. The cooling process is a reversal of the changes observedin heating. These changes are indicated on the Diagram by the

M 760

. Quenching Po'mfT

Seconds FIG. 32.DIAGRAM SHOWING HEATING AND COOLING POINTS OF STEEL.

line M, R, B, with the kink at R. The point R is of equal im-portance to the point C. It is only when a piece of steel is heatedabove the change point C that it will harden on quenching. Thepoint on cooling usually occurs at 40 degrees to 60 degreesC. below that observed on heating.The exact temperature, at which the critical change points occur,depends upon the chemical composition of the steel. Generallyspeaking, the special elements, which are introduced into aircraftsteels, have a tendency to lower their position. These changepoints have a direct relation to the physical changes which it isdesired to obtain in the material after treatment. Unless theheating is extended above the point C, no change will take pl^cein the structure of the material.

31. The "Burning" Of Steel. In the forging operation thesteel should be heated to a temperature well above the point C,say about 200 degrees to 250 degrees above it ; but, on no accountmust the temperature of the steel be allowed to extend beyond this,otherwise there is a great liability of burning the metal. If themetal is burnt it wrill become extremely " red short," i.e., itcrumbles or cracks on the edges under mechanical pressure. Ametal that has been burnt, through careless heating, is useless forany purpose whatever ; it cannot be brought back to the normalcondition by any means. It should be scrapped and remelted.Steel is burnt by being heated to a temperature nearly approach-

49


56/128

ing the melting point. The burning is caused by the evolution ofthe occluded gases. The atmospheric oxygen fills up the porespreviously occupied by these gases, and, at the same time, com-bines with some of the carbon of the steel. The crystals, orgrains, are oxidized, and are, therefore, mechanically separated bya layer of iron oxide. This explains the extreme brittlenesspossessed by steel in this condition.To illustrate this point practically, an attempt could be made toweld, without the use of a suitable flux, two pieces of steel whichhave been heated in an oxidizing atmosphere. It will be evidentbefore the experiment has proceeded very far that the task is ahopeless one, for no amount of hammering, or squeezing, willcause the adjacent surfaces to cohere.The case of burnt steel is similar. The oxidized crystals lackthe power of cohesion, and all subsequent hammering fails to weldtogether adjacent crystals. Hence the impossibility of restoringburnt steel. After the critical point is reached, each increment ofheat corresponds to an increased coarsening of the grain. Thefinal structure depends upon the maximum temperature, and themethod of cooling. If a steel is forged from a high temperature,the action of the working breaks up all crystallization ; if the workis finished near the critical point the steel will possess a compara-tively fine structure. On the other hand if a steel is allowed tocoo] undisturbedly (i.e., without any mechanical working), from atemperature greatly exceeding the critical point, the resultingstructure will be coarse.

Overheating may be caused by soaking the steel too long in thefurnace, even though the temperature may be but slightly abovethe critical point. The coarsening effect will be greatly intensifiedif the soaking is followed by slow cooling, as in annealing.Taking into consideration the size of the ingot and the size ofthe finished bar, the heating should be arranged so that the finish-ing temperature is just slightly above the critical point, i.e., from680 degrees to 750 degrees C., according to the chemical compo-sition of the material.

It must be admitted, however, that these instructions presentcertain difficulties, because it is obvious that, in the case of largeobjects, if the finishing temperature is such that the outside tem-perature of the object is just slightly above the critical point, thetemperature of the interior will exceed this point. During the slowand undisturbed cooling to atmospheric temperature, the structureof the interior will, therefore, coarsen. If, on the other hand, thefinishing temperature of the interior is near the critical point, theoutside will be below the point. And if working be carried toofar, the outside portion will be structurally distorted, and will con-sequently suffer. The effect of uneven heating in a round bar ofnickel-chrome steel is illustrated in Fig. 33. In the rolling opera-tion the billet was not heated right through, with the result thatthere was a ff crushed centre/' which is clearly visible in the illus-tration.

It sometimes happens that one end of the object attains its heatmuch quicker than the other. This difficulty can, generally, beovercome by turning the object round in the furnace. Good results50


57/128

cannot be secured by commencing the work when one end of theobject is visibly hotter than the other. If the work cannot be donein one heat without causing the finishing to be done at too low atemperature, the metal should be partly reduced in size, and thenreturned to the furnace to be reheated before finally finishing.The second reheating temperature should not be as high as thefirst. It should be understood that steel deteriorates with too fre-quent heatings; therefore the number of heatings should be mini-mized. The metal should be reduced in size as much as possibleduring each heat.As can be expected, the struc-ture of a hot-worked steel dependsupon the initial temperature andthe finishing temperature. If thefinishing temperature is butslightly above the critical pointC, the structure will be compara-tively fine; but if mechanicalwork is discontinued at a tem-perature greatly exceeding thecritical point, then the structurewill be relatively coarse. It isevident that, through forgingalone, it is impossible to impartan absolutely uniform structureto steel objects of large crosssection. The lack of uniformity inthe structure of the material re-quires that it should be normal-ized or annealed before it is sentforward for machining operations.

32. Annealing. The purpose of annealing is to remove all theinternal stresses that have been induced by mechanical work, andalso, at the same time, to change the heterogeneous structure intoone possessing a uniformly fine grain. Annealing minimizes theliability of distortion in the hardening operation, for, when heated,the steel is then able to expand more easily. After annealing themetal possesses the maximum softness and ductility, and is in thebest possible state for machining. The annealing operation con-sists essentially of three stages, viz., heating up, soaking andcooling. For a given steel it is desirable always to anneal at thelowest possible temperature. It is necessary to heat just abovethe critical point C (Fig. 32), and to keep the object, as near aspossible, at that temperature throughout the process.To ensure the refinement of the grain, the object should be keptat a constant annealing temperature until the desired structuralchanges have taken place in the steel. To bring about the desiredeffect the temperature should be maintained i to 4 hours accord-ing to the bulk of the object. On the other hand, the metal shouldnot be kept at the maximum temperature any longer than isnecessary for the heat to get right through it. Maintaining it, fora long time, at the temperature defeats the purpose of annealingby causing a coarsening of the grain.

51

FIG. 33.CRUSHED CENTRE IN ROUNDNICKEL CHROME STEEL BAR.


58/128

33. Normalizing. " Normalizing " a metal means heating itto a temperature exceeding the upper critical point C (Fig. 32),and allowing it to cool freely in air. The temperature should bemaintained for about fifteen minutes, and should not exceed thepoint C by more than fifty degrees centigrade.

Normalizing an aeroplane fitting will improve its condition(a) By refining the structure of the steel after it has beenheated to a high temperature, as in forging, stamping, weld-

ing or brazing.(b) By releasing the stresses left by cold working, as broughtabout by forging at too low a temperature, bending, pressing,

etc.(c) By releasing the stresses left by local heating or uneven

cooling.The parts to be normalizedshould be left in the furnacelong enough to ensure thatthey have attained a uniformheat throughout. They shouldthen be removed from thefurnace and allowed to coolfreely in the air. The effect ofnormalizing may be judgedfrom the microphotographsshown in Figs. 34 and 35. Thesecorrespond to the same materialafter normalizing from 900degrees C. as the microphoto-graphs taken from the 7-inchand 7-8-inch square respective-ly, shown in Figs. 30 and 31.The physical tests correspond-ing to these microphotographsare :

FIG. 34.MlCROPHOTOGRAPH OF SECTIONCUT FROM 7 INCH FORGING .30CARBON STEEL AFTER NORMALIZ-ING. MAG. 200 DIAMS. ANDREDUCED.


59/128

contributing to a particular failure. However, since most ofthe difficulties met with are now understood, it only requiresforethought in order to adopt preventive measures to coun-teract them. It does not necessarily follow that all cases offailure are due to bad material. Failure may arise through theuse of unsuitable material; and, in quite a number of instances,it may be attributed to faulty design. For instance, sharp angularcorners are always a source of danger in the hardening shop, andshould, therefore, be avoided as much as possible.The rate of heating and cooling influences the final physicalproperties of a material. Steel articles, that are subject, to heattreatment, display some uncertainty in the uniformity of the results,which may be expected, after scientifically defining the treatment.Some objects possess thick and thin parts, consequently thethinner parts take the heat more rapidly than the bulky portions ;and in the quenching operation they also lose their heat thequickest. This often accounts for a difference in the physical condi-tion of different parts of the same article. Therefore it is obviousthat each individual object should be studied with regard to itsshape or volume in order to produce uniform results.The chief difficulties met within the hardening shop are distor-tion and cracking, both of whichmay be largely overcome byadopting proper methods of heat-ing and cooling. Distortion iscaused by uneven heating andcareless quenching. It can beavoided by slowly heating to therequired temperature, and soakinguntil a uniform heat is attainedthroughout the entire mass. Dur-ing quenching, the object shouldbe immersed in the bath vertically,in the direction of the main axis.

FIG. 35.MlCROPHOTOGRAPH OF SECTION CUTFROM 7-8TI1 INCH FORGING .30 CAR-BON STEEL AFTER NORMALIZING.MAG. 200 DIAMS AND REDUCED.

If the object possesses thickand thin adjacent parts, the thickparts should strike the liquidfirst, so as to offer large surfacecontact with the cooling medium.Provided that the material is free from original defects, cracksmay be caused, during hardening, by uneven cooling in the crosssection of the object. On quenching, the outside surface is trans-formed to the hardened state and, depending upon the thickness,the hardening influence gradually extends towards the centre, untilthe entire mass attains the temperature of the bath. If the objectis withdrawn from the bath, before the centre is quite cold, thereis a great liability of cracks forming. This is due to the contrac-tion of the still warm centre, which exerts an enormous force onthe outer surface, which, being in the hardened and brittle state,is unable to withstand the contractive force, and consequentlyfracture is caused.

53


60/128

Many misleading ideas are current with regard to the correctmethod of hardening steel. For instance, one branch of theoristsadvocate that the steel should be heated just through the criticalrange, and then cooled suddenly while the temperature is stillrising. This method, they state, confers the finest possible struc-ture on the material. Moreover, unfortunately, many practicalsteel-hardeners think that it is absolutely essential to quench themetal from the highest heat attained, and accordingly this methodof procedure is the one largely followed in works practice. Herein

TABLE IITable showing the Brinell Hardness Numbers obtained afterQuenching from a Falling Temperature.


61/128

the heat was still rising. The furnace was allowed to cool slowly,and the consequent specimens were quenched when the tempera-ture had fallen to the required degree.

In works practice, the object should be slowly heated and soakedat the correct hardening temperature, and then allowed to cool,say 50

or 60 degrees in the furnace before it is finally quenched.35. Tempering. " Tempering " means heating the metal,

after hardening, to a temperature not exceeding the critical pointR. Tempering reduces the hardness, and increases the toughness,to a degree depending on the temperature used.

55


62/128

CHAPTER VCASE-HARDENING

36. Importance of Case-Hardening. Case-hardening is oneof the most important heat-treatment operations. It affords themost satisfactory method of fulfilling the requirements of parts,such as camshafts, cams, gudgeon-pins, gears, etc., which musthave an intensely hard surface to resist wear, and, at the sametime, must be fairly tough and ductile to withstand shocks. Amedium high-carbon steel, hardened in the ordinary way, can bemade to possess the requisite hardness on the surface, and thusbe capable of withstanding hard wear. But in this conditionit will be too brittle to resist shock or excessive vibrations. Ifthis same steel be tempered, with the object of making it tough,the hardness of the surface will be lowered, and the material will,consequently, be too soft for the purpose. If, on the other hand,a mild steel be heat-treated, it will perhaps possess the requisitetoughness, but the surface will not be hard enough to resist wear.Parts requiring, as their ideal features, a good hard surface anda tough interior, should be submitted to the case-hardening pro-cess. By employing a material low in carbon, and heating it incontact with a suitable carburizing agent, and then, by subjectingit to a series of heatings and quenchings, the desired character-istics will be obtained, viz., a hardened high-carbon exterior, anda low-carbon toughened interior. The case-hardening process isdivided into three separate stages, viz., the carburizing stage, thefirst quenching to refine the core, and the second quenching torefine the case.

37. Carburizing Compounds, The carburizing operation maybe carried out by either solid, liquid or gaseous "compounds." Thesubstance by which carburization is effected must be such as easilygives up carbon. Carburizing substances can be divided intothree classes : firstly, solids containing a large proportion of" fixed carbon," or in which the carbon is present as a gaseoushydrocarbon, or in a form from which gaseous hydrocarbon canbe liberated; secondly, liquids such as molten cyanides; andthirdly, gaseous compounds such as carbon monoxide, hydro-carbons and cyanogen. For industrial purposes solid and liquidcompounds are in general use.

56


63/128

A liquid bath of potassium cyanide can be used when a rapidrate of penetration is required, and the part to be case-hardened isonly very thin in section. As liquids only give a superficial case,they are unsuitable for most classes of work. The molten cyaniderapidly deteriorates and loses its carburizing effect. Therefore itis necessary constantly to add new material to make up for theinertness of the old. Cyanide compounds are extremely dan-gerous to use, since the fumes given off by the molten bath consti-tute a deadly poisonous vapour. Hence, when these compoundsare used, adequate provision should be made effectively to carrythe fumes away from the operator. Fig. 36 shows an apparatussuitable for carrying out the carburizing process wTith moltencyanides.

38. Selecting a Carburiz-ing Compound. From themany solid compounds in useit is a difficult matter to selectthe one which is the most suit-able for general purposes. Agreat deal is usually said bythe makers of case-hardeningcompounds concerning thevaluable properties possessedby their own particular brands,and of their inherent value overall other mixtures. The selec-tion of a compound dependson the nature of the work tobe handled. A mixture that issatisfactory for one class ofwork may be entirely unsuit-able for another. On accountof their high cost, shops deal-ing with a large quantity ofwork will not favour the useof those brands which are richin volatile hydrocarbons. Al-though these compounds arevery active when new, and givea quick penetration, they haveonly a comparatively short life,and are expensive in the initialcost. Moreover, when a gooddepth of case is desired anextremely active agent isunsuitable on account of itsshort life. The effectivenessof the carburizing agent is notaltogether dependent on itsspeed of penetration : themost important factor is theamount of surface carbon that

FIG. 36.GAS FURNACE SUITABLEFOR HEATING MOLTEX

CYANIDES.

57


64/128

will be absorbed from it by the steel. Wood charcoal is the chiefconstituent of carburizing mixtures which are used in ordin

1920 metalsinaircraft00hanbrich

Documents