ELECTROS LAG CASTING OF VALVE BODIES by DEEPAK GUPTA B.Tech., Indian Institute of Technology, Kanpur, India, A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Metallurgical Engineering We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA January 1982 (c) Deepak Gupta, 1982 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Da te OOW- f8- , t^S2~-3E-6 (2/79) ii ABSTRACT The Electroslag Casting (ESC) process has been widely used in the Soviet Union for the production of high quality steel castings. This work presents the results of an examina tion of the ESC process for the production of simple shaped valve bodies. Stainless steel (AISI 316 and ACI CF-8M) and low alloy steel (AISI 4340) valve bodies were made at U.B.C. and tested by non-destructive and destructive methods. It is concluded that this technique offers distinct quality and production advantages and the properties easily meet the required ASME/ASTM specifications. Therefore they are equi valent to or better than the conventional castings and forc ingsHowever, there may be difficulties in reconciling the method with present code qualification requirements. TABLE OF CONTENTS Abstract Table of Contents List of Tables List of Figures — Acknowledgements • Chapter 1 INTRODUCTION 2 PROCESS DESCRIPTION AND CHARACTERISTICS 2.1 Electroslag Remelting Process Characteristics .. Which Affect the Properties of the Materials ... 2.1.1 Characteristics Related to Chemical Refining 2.1.2 Characteristics Related to Solidification structure 3 PRESENT WORK 3.1 Furnace Design 3.2 Valve Mold Design 3.3 Melting Procedure 3.4 Selection of Slags for ESC Process 4 EVALUATION OF ESC VALVE CASTINGS A. STAINLESS STEEL VALVE CASTINGS 4.1 Remelting Log for Stainless Steel ESC Valves ... 4.2 Non-Destructive Testing (NDT) 4.2.1 Dye Penetrant Test 4.2.2 Ultrasonic Test 4.2.,3 Radiography Test 4.3 Destructive Testing (DT) 4.3.1 Macrostructures 4.3.2 Microstructures 4.3.3 Delta Ferrite Measurements 4.3.4 Interdendritic Microsegregation 4.3.5 Composition Analysis and Macrosegregation ... 4.3.6 Problems With Alloying During the ESC Operation 4.3.7 Problems With Using Cast Electrodes in ESC Operation 4.4 Mechanical Properties 4.4.1 Tensile Testing B. LOW ALLOY STEEL VALVE CASTINGS 4.5 Remel.ting Log for AISI 4340 ESC Valves 4.6 Non-Destructive Testing (NDT) 4.6.1 Dye Penetrant Test 4.6.2 Ultrasonic Test 4.6.3 Radiography Test 4.7 Destructive Testing (DT) 4.7.1 Macrostructures 4.7.2 Sulphur Prints 4.7.3 Interdendritic Microsegregation 4.7.4 Composition Analysis and Macrosegregation ... 4.7.5 Heat Treatment and Microstructures 4.7.6 Dimensional Measurements 4.8 Mechanical Properties 4.8.1 Tensile Testing 4.8.2 Impact Testing 5 OTHER TRIALS AND FUTURE WORK 6 SUMMARY AND CONCLUSIONS REFERENCES TABLES^ FIGURES APPENDIX 1 - ASME/ASTM Specifications V LIST OF TABLES Tables Page I Remelting Log for Stainless Steel ESC Valves 83 II Ferrite Numbers of Stainless Steel Castings 84 III Average Ferrite Numbers of Stainless Steel Castings Determined by Magne-Gage and Schoefer's Diagram 85 IV Interdendtitic Microsegregation Ratios of CF-8M Castings .. 86 V Chemical Composition of Valve No. 5 (CF-8M) 87 VI Chemical Composition of Valve No. 6 (316) 87 VII Chemical Composition of Valve No. 7 (316+Cr) 88 VIII Chemical Composition of Valve No. 9 (CF-8M) 88 IX Chemical Composition of Valve No. 10 (316+Cr+Mo) 89 X Chemical Composition of Conventional Casting (CF-8M) 89 XI Percent Composition of Cr and Mo in Areas 1 and 2 and the Parent Valve Casting No. 10 90 XII Tensile Properties of Valve Nos. 5, 6 and 7 90 XIII Tensile Properties of Valve No. 9, Conventional Casting and ASME/ASTM Standards for AISI 316 and ACI CF-8M 91 XIV Tensile Properties of large specimens from Valve Nos. 6 and 9 and small specimens from a large specimen 92 XV Remelting Log for AISI 4340 ESC Valves 93 XVI Interdendritic Microsegregation at the Centre of AISI 4340 ESC Valves 94 XVII Chemical Composition of Valve No. 3 (4340) 95 XVIII Chemical Composition of Valve No. 8 (4340) 96 XIX Chemical Composition of Valve No. 13 (4340) 97 XX Chemical Composition of Valve No. 14 (4340.) 9Tables vi Page XXI Sulphur Contents of the AISI 4340 Electrode and ESC Valves. 98 XXII Dimensional Measurements on Valve Nos. 11, 12 and 13 98 XXIII Tensile Properties of AISI 4340 Electroslag Cast Valves ... 99 XXIV Tensile Properties of Conventional AISI 4340 Hot Rolled Bar 100 XXV Transverse Tensile Properties of Air-melted and Vacuum-arc Melted AISI 4340 Steels75 10XXVI Longitudinal Mechanical Properties of Bar Stock Made From Remelted AISI 4340 Steel75 100 XXVII Mechanical Properties of ESR 4340 Material in the Trans- :il verse Direction Heat Treated to Different Strength Levels^ ... 101 XXVIIIa Tensile Properties of Large Specimens from ESC Valve No.8 (AISI 4340) 10XXVIlib Tensile Properties of Small Specimens from Large Specimen from ESC Valve No. 8 (AISI 4340) 101 XXIX FATT Values Estimated from Ductile-Brittle-Transition Characteristics of the ESC and Electrode Material 102 VI 1 LIST OF FIGURES Figure Page 8 1 Electroslag cast products 103 2 'YOZO' process of Mitsubishi Heavy Industries Ltd.11'12... 104 3. Schematic of the ESC process 105 4 Fracture toughness of ESC AISI 4340 steel containing varying amounts of sulphur^ 106 5 Electroslag casting installation at U.B.C 107 108 109 6 Orthogonal views of the valve mold segments 7a Plan views of the valve mold segments 7b Sections through the valve mold segments 110 8 Assembled valve mold Ill 9 Slag melting furnace HI 10 Solidified slag skin on the casting 112 no IT Phase diagram for CaF^AjZ^-CaO system 113 12 Stainless steel ESC valve H4 13 Longitudinal holes in the castings due to moisture H5 14 Schematic of the sectioning procedure of the valves H6 15 Macrostructure of Valve No. 5 (CF-8M) 16 Macrostructure of Valve No. 6 (316) H7 17 Macrostructure of Valve No. 9 (CF-8M) 118 1.8 Macrostructure of Valve No. 10 (316+Cr+Mo) 119 19 Macrostructure of conventional casting (CF-8M) 120 20a Miicrostructure of CF-8M ESC valves before heat treat ment (etchant - oxalic acid) 121 20b Microstructure of CF-8M ESG valves after heat treat ments (etchant - oxalic acid) 21 Microstructure of as-cast ESC Valve No. 5, 6 and 7 (etchant - oxalic acid) 22 Microstructure of heat-treated ESC Valve No. 5, 6 and 7 (etchant - oxalic acid) 23 Microstructure of CF-8M specimen heated to 850°C (etchant - KOH) 24 Microstructure of heat treated CF-8M conventional casting (etchant - KOH) 25 Microstructure of Valve No. 9 in as-cast condition (echant - KOH) 26 Microstructure of Valve No. 9 in heat-treated condition (etchant - KOH) 27 Microhardness indentations on sigma and austenite phases.. 28 Schoefer's diagram for determination of ferrite numbers of stainless steel castings66 29a Variation of Cr, Ni and:Mo across the dendritic . direction in Valve No. 9 at the edge 29b Variation of Cr, Ni and Mo across the dendritic '',*; direction in Valve No. 9 at the centre.. 30a Variation of Cr'v.;Nf and ;Mo across the 'dendritic : '"' v: ; direction in conventional casting at the edge 30b Variation of Cr, Ni, and Mo across the dendritic direction in conventinal casting at the mid-radius 30c Variation of Cr, Ni, and Mo across the dendritic direction in conventional casting at the centre.. 31 Composition variation in Valve No. 5 (CF-8M) 32 Composition variation in Valve No. 6 (316) 33 Composition variation in Valve No. 7 (316+Cr) Figure 34 Composition variation in Valve No. 9 (CF-8M) 35 Composition variation in Valve No. 10 (316+Cr+Mo) 36 Composition variation in conventional casting (CF-8M) 37 SEM photographs of agglomerated ferro alloy powder in area 1 in Valve No. 10 38 SEM photographs of agglomerated ferro alloy powder in area 2 in Valve No. 10 39 EDXA plots of agglomerated ferro alloy powder in areas 1 and 2 and the parent metal (Valve No. 10) 7. '. 40a EDXA plots of inclusions in areas 1 and 2 and Fe-Cr powder 40b EDXA plots of inclusions in areas 1 and 2 and Fe-Cr powder 40c EDXA plots of inclusions in Fe-Mo powder 41 Macroporosity in the centre of the CF-8M electrode 42 Macrostructure of the electrode piece that dropped in Valve No. 9 43 Macrostructure of the CF-8M electrode dropped 44 Inclusions in the electrode piece dropped in Valve No. 9 and electrode tip (optical photographs) 45 Inclusions in the electrode piece dropped in Valve No. 9, electrode tip and the parent casting (Valve No. 9) 46a EDXA plots of inclusions in the electrode piece dropped in Valve No. 9 and the electrode tip 46b EDXA plots of inclusions in the parent casting (Valve No. 9) 47 Schematic to explain the peculiar pool profile observed in Valve No. 9 48 Macrostructure of the peculiar pool profile observed in Valve No. 9 Figure 49 Schematic of the tensile specimens used .. 50 Variation of tensile properties with ferrite number of stainless steel castings 51 Photograph of the deformed and fractured areas of small tensile specimen from Valve No. 6 52 Photograph of the deformed and fractured areas of large tensile specimen from Valve No. 6 53 AISI 4340 ESC valve 54 Cracks in Valve No. 14 revealed by dye-penetrant test 55 Radiographs of Valve No. 3 56 Macrostructure of Valve No. 3 Top - HC£ etch Bottom (NH4)2 S04 etch 57 Macrostructure of Valve No. 8 (HC£ etch) 58 Macrostructure of Valve No. 13 (HCP. etch.)'-.,. 59 Macrostructure of Valve No. 14 (HC£ etch) (top part etched for a longer time) 60 Macrostructure of the transverse section of a mild steel ESC valve (HC£ etch) 61 Dendritic structure of Valve No. 3 62 Sulphur prints of Valve No. 3 63 Sulphur prints of Valve No. 8 64 Variation of Cr, Ni, and Mo across the dendritic direction in Valve No. 8 65 Variation of Cr, Ni and Mo across the dendritic direction in Valve No. 13 . — 66 Variation of Cr, Ni and Mo across the dendritic direction in Valve No. 14 67 Composition variation in Valve No. 3 Figure 68 Composition variation in Valve No. 8 69 Composition variation in Valve No. 13 70 Composition variation in Valve No. 14 71 Machined AISI 4340 ESC valve 72 Separated surfaces along a crack in Valve No. 14 73 Hardness variation in heat-treated Valve No. 13 74 Hardness variation in heat-treated Valve No. 14 75 Microstructures AISI 4340 valve in as-cast and heat-treated conditions; electrode in heat-treated condition 76 Microstructure of AISI 4340 ESC Valve No. 13 in heat-treated condition 77 Fractographs of large tensile specimens of AISI 4340 from ESC valve No. 8 78 Orientation of the charpy specimens and the notch in Valve No. 8 , 79 Ductile brittle transition characteristics of Valve No. 3 and the electrode 80 Ductile brittle transition characteristics of Valve No. 8 81 Ductile brittle transition characteristics of Valve No.13 82 Optical fractographs of charpy specimens from valve No. 8 tested at different temperatures and orientations a - TT b - TL c ~ E d - E ..... 83 Optical fractographs of charpy specimens from the AISI 4340 electrode 84 SEM fractographs of charpy specimen a - ridge area b - micro-cracks .. 85 SEM fractographs of different regions of a charpy specimen xi i Figure Page 86 Soviet electroslag cast valve 193 87 Schematic of the methods used for making hollow ESC valves 194 88 Electroslag casting of a flange for use in valve bodies .. 195 89 ESC valve with the welded insert 1990 Macrostructure of the ESC valve with the welded insert ... 196 ACKNOWLEDGEMENTS xi i i I would like to thank sincerely my research supervisor, Professor Alec Mitchell, for his friendly assistance and guidance throughout the course of this research project. Thanks are extended to my fellow graduate students of the Department of Metallurgy for their helpful suggestions. The assistance of the technical staff of this department, in particular Mr. Gus Sidla, is greatly appreciated. And finally, I am thankful for the financial assistance of the Natural Sciences and Engineering Research Council of Canada and of the American Iron and Steel Institute. Chapter 1 1 INTRODUCTION With the growth and expansion of the aerospace and nuclear industry, the demand for specialty; metals with ex ceptionally high standards of quality and reliability has in creased tremendously. The specifications for the components are becoming more and more rigid. The result is, as might be expected,^..- a high rate of rejection of conventionally made products which fail to pass rigorous testing prior to instal lation and fail when in service. Although the Vacuum Arc Remelting (VAR) process was conceived as early as the mid-nineteenth century, it has only been used extensively in the last thirty years. Electroslag Remelting (ESR) process, on. the other hand, is relatively new. Over the last fifteen years, ESR has been established as an industrially-viable process for the production of high quality material.in a wide composition range. The common objective behind the development of both these remelting processes was that of producing high quality material by chemical refining and con trolled solidification. The possible high cost of remelting is offset by the processes' advantages of increased service life of the components, very low rejection rates, and high overall yield. 2 Initially, the ESR process was widely used in U.S.S.R. for upgrading the air-melted quality material. Then the use of ESR was extended to forging applications and the process gained prominence in other countries; also. During recent years it became evident that the cast ESR material was equi valent to or better than the conventional castings and forg-ihgs. This led to speculation as to whether or not the ESR process can be extended to produce shapes very close to the final cast or forged product (near net shapes). The next step in this logical sequence was the emergence of the Electro-slag Casting (ESC) process. Generally, the conventionally cast components are associ ated with defects such as shrinkage, porosity, large non-metallic inclusions, segregation and hence have poor and anisotropic mechanical properties. Although with recent cast ing techniques, such as squeeze-casting, it is possible to produce better quality castings, these are limited to a small scale only and involve heavy capital expense. The hot working processes ( f org img :and rol 1 ing) can be applied to eliminate the casting defects and produce a better quality material, but the increase in anisotropy of the mechanical properties and extra cost makes this step unattractive. However, with the ESC process it is not only possible to produce large castings with the same ease as the small ones, but they are free from most of the defects mentioned above and are manufactured at a comparatively low cost. 3 The E.O. Paton Institute in Kiev, U.S.S.R. has extensively used the ESC process to produce components for a wide variety of applications. These range from dentures weighing a few grams to very large valve bodies, rolls, pressure vessels, 1-2 crankshafts:, ; gears, etc. weighing several tonnes (see Figure 1). Numerous examples of these are published in the 3-23 literature. The production of fuel-hand!ing components like valves, pumps, and steam fittings for boiling water reactor nuclear power plants and the petrochemical industry 3-9 has been reported by the Soviets on several occasions. ESC tubes and fittings for the petrochemical industry have been manufactured by the Mitsubishi Corporation of Japan.10-12 They have used the ESC method (YOZO) to produce reformer tubes in HK alloys, (see Figure 2). Tubes with varying section, elbows, oval sections, header tubes and some 'I' beam sections have been produced.11 Another area of ESC application is in the manufacture of crankshafts, rolls and gears. The production of diesel engine crankshafts up to 160 tonnes in weight has 4-7 9 been reported ' and they are presently in use in the Soviet Union. Heavy duty kiln-ring gears in sizes up to 150 tonnes 18 13 for 6-7 meter diameter kilns are reported to be in service. ' ' Raton also reports' of the manufacture of ESC rolls for use in the rolling of steel.5_8>14 Other reported ESC applications include die blocks in 4 alloys similar to Hll and H13, cast close to the finished shape, ' hobbing cutters made from Tl and M2 tool steels 17 18 cast by ESC to the semi-finished shape, gun tubes, pre-o cision dentures in austenitic stainless steel, and nozzles cast on pressure vessels. It is evident from the above list that ESC indeed offers a promising prospect for producing finished shape components without any hot or cold working operation. It is also obvious that most of reports available are of Soviet origin and little or no work has been done in this field in North America. Hence it is of considerable importance to investigate this promising aspect of the Electroslag Remelting process and to determine if the properties of the ESC products qualify under the ASME/ASTM codes, given North American economic manufactur ing constraints. Chapter 2 5 PROCESS DESCRIPTION AND CHARACTERISTICS The electroslag remelting process is similar to the vacuum arc remelting process in that both processes use a solid consumable electrode and melt it through a high-temper ature process region into a water-cooled mold. However, they differ in the heating mechanism. Unlike WAR, where heating is through an arc, the ESR process uti1izes resistance heating of a liquid slag between the electrode and the ingot surface. Detailed accounts of the operation of the ESR process are avail-24-29 able in literature. The concept of the ESC process is the same as the ESR process except that the ESC process uses a shaped water-cooled mold. This is illustrated schematically in Figure 3. A consumable electrode of desired composition is made as one pole of a high current source and a water-cooled copper or aluminum:- shaped mold is the other pole. The Joule heat generated in the slag pool melts the electrode tip and molten metal drops fall through the liquid slag into the molten metal pool from which the casting solidifies in the water-cooled mold in a directional, progressive manner. Ef fective refining of the metal with respect to sulphur, phos phorus and non-metallic inclusions takes place during the formation of the metal drop, as it traverses through the slag pool, and in the liquid metal pool. A continuous shell of 6 solid slag forms between the mold and the solidifying casting and this provides smooth surfaces to the casting. A sound casting ,i s built up progressively by a nearly vertical solidi fication pattern. All this effectively eliminates macrosegre-gation; brings mierosegregation to a minimum; and removes poro sity, shrinkage, hot cracks and other defects related to con ventional castings. 2.1 Electroslag Remelting Process Characteristics Which Affect the Properties of the Materials Almost all the existing 1iterature on ESR, has, at one point or another, extolled the virtues of the process. Of these the ones which affect the mechanical properties can be broadly classified under the following two categories: those related to chemical refining. those related to solidification structure. 2.1.1 Characteristics Related to Chemical Refining The ESR process has earned its name from the inherent re fining characteristics. The most important is the removal of large quantities of sulphur, even when the level of this is 30 31 low in the starting material. ' Very low sulphur levels can be obtained by using highly basic slags. It is well esta blished that one of the most effective ways of obtaining good and reasonably isotropic mechanical properties is through 7 32-35 sulphur control. Figure 4 shows the fracture toughness results of a series of ESR 4340 steels containing varying amounts of sulphur. Since sulphides are initiation sites for fracture in many cases, control of their shape and size re sults in uniform fracture characteristics in three primary directions. It is noteworthy that with the recent advances in ladle steelmaking technologies, it is possible to achieve low sulphur levels and sulphide inclusion shape control. Hence sulphur control is no longer considered to be an important factor deciding for or against the ESR process. However, although lowering of the overall sulphur level of the steel provides the most significant improvement, de-sulphurisation is only the first step. Removal of other in clusions and detrimental residual impurities, and controlling the shape and distribution of sulphide inclusions are also of considerable importance in obtaining the optimum mechanical properties. Although many references are available in the literature which deal with the type, size, content and distri bution of inclusions in ESR material, very few refer to the actual mechanism by which the non-metallic inclusions are removed from the material during electroslag remelting. As the casting is solidified in a water-cooled metal mold, the chance of exogenous inclusions being present in the remelted material is very low. However, indigenous inclusions of either de-oxidation (formed by dissolved oxygen and the deoxidants) 8 and/or solidification (formed by precipitation during solidifi cation) are present in ESR material. Some of the early work shows that the electrode inclusion content affects the in-elusion content of the ingot, but it is well accepted now, 37 as Mitchell points out, that virtually no ingot inclusions are identifiable as unreacted electrode relics.' In the liquid metal pool there are few inclusions present, unless very high deoxidation rates are used. The liquid has only dissolved oxygen and deoxidants in near equilibrium with the slag. When the liquid metal freezes, inclusions characteristic of the oxygen and deoxidant content of the liquid are nucleated and OC Of) O ~l grown in the ingot liquid-solid zone. ' ' This process gives rise to a fine background of small, globular and dis persed inclusions. The total volume fraction in ESR is much lower than the best air-melting practice. Generally, inclusions have a greater effect on non-uniform deformation (percent reduction in area) than upon uniform de formation (percent elongation). If the inclusions have an anisotropic shape (e.g. elongated sulphide inclusions in rol led material), ductility will be anisotropic. Therefore, the non-metallic inclusion content of remelted steels is of prime importance in determining their mechanical behaviour. 2.1.2 Characteristics Related to Solidification Structure Contrary to the conventional casting processes in which 9 melting and casting represent two separate and different opera tional steps, the ESR process is characterized by addition of heat during the solidification step. Hence the temperature gradients and the solidification rate can be controlled to esta blish a progressive and directional solidification condition (obtained by high thermal gradients and low growth rates). Such slow melt rates are possible firstly due to the insulating character of the slag pool which levels off the thermal gra dients across the pool and secondly due to the formation of the thin slag skin between the mold wall and the metal. These soli dification conditions lead to a structure which is practically free of porosity, shrinkage, hot-cracks, etc. This also ef fectively eliminates macroscopic inhomogenities such as banded structure or freckles, macroscopic spot segregation, and brings microsegregation to a minimum. Although low melt rates lead to a fine dendritic structure, mvcrosegregation is pronounced as the system deviates from equi librium solidification. Also very high melt rates lead to an equiaxed structure with increased microsegregation. An optimum melt rate leads to least microsegregation and hence good mechani-3 8 cal properties. In conclusion, it can be said that the enhanced mechanical properties and isotropic., behaviour of the ESR material are due to good chemical refining and solidification structure. Chapter. 3 10 PRESENT WORK In the latter part of 1974, a program was initiated among Dr. Alec Mi tchell , Professor of Metal lurgy U.B.C, Westi nghouse Electro-mechanical Division (WEMD) and Dr. Boris Medovar, Director of ESR Technical Development at Paton Institute, Kiev, U.S.S.R. A 10-inch ESC valve body was made in the U.S.S.R. and sent for evaluation at U.B.C. and WEMD. Evaluations of destruc tive and non-destructive testing of the valve body are avail able.39'40 The results indicate that the casting quality failed chemically and mechanically to meet ASME requirements as origi nally predicted by the Soviet workers. Hence to study in-depth the properties and the process viability of ESC products, a simple and inexpensive ESC furnace was installed at U.B.C. The furnace is capable of making cast ings up to one tonne. The furnace has made valve-body castings in alloy and stainless steels from both wrought and cast elec trodes . 3.1 has Furnace Design Detailed account of the design been reported previously.41 and operation of this furnace 11 The mechanical part of the design consists of a water-cooled electrode holder carriage which can move vertically using aluminium rail guides. These guides are fixed to the inside of a vertical I-beam framework. The carriage is suspended from a single chain drive having a variable speed reductor and this in conjunction with a speed controller gives a range of electrode speeds from 0-163 mm/min (see Figure 5). The electrical part consists of two step down transformers. The first one is rated at 250 KVA with 12500 V input and 600 V output. The second transformer is single phase dry type and is also rated at 250 KVA with input voltage of 600 V and output voltage variable over a range of 25V - 60V in steps of about 2.5 V. The operating parameters can be monitored from the control panel which not only shows the primary and secondary current and secondary voltage but also has a digital counter to measure the electrode travel. Also, a multichannel thermometer shows the mold cooling water temperatures. 3.2 Valve Mold Design Smooth surface finish, flat tops and bottoms, controlled vertical solidification and shaped products are some of the discussed unique features of ESR ingots. All of these are directly attri butabl e to or are strongly i nfl uenced by mold design 12 and operating parameters. The optimum mold design is associa ted with the manner in which the mold assembly dissipates the heat primarily from the slag layer, molten pool and solidified ingot casting. ESR mold design has been discussed to some ex-25 42 tent-.by Mi tchel 1 and Cremesio et al . Most of the ESR molds used are made out of copper because of its high thermal conductivity. This adds substantially to the overall cost of the product. Fabrication of a copper mold also presents a problem as the metal has to be formed and welded into the mold shape. This is particularly difficult for ESC molds with contoured sections. Since the inventory costs and fabrication make copper molds unattractive, a cast-to-shape mold of less costly aluminum was considered to be a viable alternative. Gear castings using a cast aluminum mold were made success fully in U.B.C. The water channel positions and dimensions in the gear mold were roughly calculated to give a maximum hot-43 face temperature of 300°C during the ESC process. Following this, an aluminum mold for casting of valve bodies was manu factured . The valve mold was cast out of aluminum using a steel pattern. The orthogonal views of the valve body are given in Figure 6. Accordingly the complete valve mold was divided into 13 four segments - two plate molds for the flat sides of the valve and two molds with cupped depressions to accommodate the inlet/outlet sections of the valve. The sectional draw ings of these segments are given in Figure 7a-b. Figure 8 shows the assembled valve mold. Besides this, two round section molds were used - one on top and the other at the bottom of the valve mold. 3. 3 Melting Procedure The valve mold is placed between two round molds and the whole assembly is placed on a water-cooled copper base plate. To prevent any damage to the base plate due to arcing with the electrode at the initiation of the process, a starter block is used. There are two different practices employed to ini tiate the process. One is known as 'molten-slag' start and the other is the 'dry-slag' start. Both the processes were tried and they worked equally well. In the 'molten-slag' start the main slag component, i.e. CaF^, is melted in a slag melting furnace (see Figure 9). This furnace has a graphite crucible and employs a single-phase graphite electrode system. CaF2 is added to the 'furnace and power is applied to initiate an arc which forms a small pool of molten CaF2 and then the heating is through resistance heat ing of the slag. When all the slag (CaF2) has melted, it is poured into the water-cooled mold and the electroslag furnace 14 is energized. After this the remaining slag components are added slowly. In the 'dry-slag' start, a bed of turnings of a metal similar to the electrode is placed between the starter plate and the electrode. Alternatively a pellet of compressed mix ture of CaF2 and metal turnings can be used. Either one of these is used and the electrode is lowered such that it pres ses the turnings against the starter plate. Then the dry CaF is poured into the surrounding annulus between the electrode and the mold walls. When the furnace is energized, power is applied across the short-circuit current path established by the turnings .under the electrode. This causes melting of the slag and continues till a large enough volume of molten slag is generated to sustain resistance heating. After all the CaF2 is melted, the remaining slag components (CaO, Ai^O^* L^O.^ etc.) are added. The electrode is lowered as it melts and the mold is slowly filled up with resolidified metal. At the end of the run, power is terminated and the electrode is withdrawn from the mold, the casting being taken out after about thirty min utes. The castings have a smooth surface due to a thin, con tinuous shell of solid slag which forms between the mold and the solidifying metal (see Figure 10). 15 The main operating parameters established during the pro> duction of valve castings were as follows: operating voltage 36-39V primary current 200-260 A secondary current 3000-4000 A - melt rate 0.8-1.0 kg/min. 3.4 Selection of Slags for ESC Process The slag pool is perhaps the most important component in the electroslag remelting process. It is required to 'fill a number of simultaneous functions - it must act as a heat source, heat-transfer medium, metal container and metal re-44 ... finer.' Therefore the correct choice of slag composition is of prime importance for successful ESR operation. The choice rests upon the following: a) heat generation and transfer characteristics of the slag - (electrical conductivity, thermal capacity and thermal conductivity). b) slag phase properties - (vapour pressure, liquidus tem perature, viscosity, surface tension and density). c) metal refining characteristics (chemical composition). The more important of the above points are discussed below with reference to slag selection. 16 The ternary CaF2 - CaO - M^O^ system is the most impor tant and universally applicable slag system in the field of ESR. Fluoride generally decreases the solidus and liquidus temperatures and viscosity, and increases electrical conduc tivity of CaF2 + A«-2U3- It has been observed that relatively small additions of A^Og to pure CaF2 produces a substantial decrease in the conductivity of the liquid. The work of 45 Mitchell and Cameron showed that this occurs because the fluoride ion contribution to the total ionic mobility is re-3-duced by complexing it in A&OF,, and A£02F2 . This would be through the reaction A*2.03 + 4F" > A£02F2" + A£0F~ This effect is also seen in the addition of silicates and of rare earth oxides. However, addition of CaO replaces the fluoride part of the complex ions by oxide ions thus releas ing fluoride ions as charge carriers, e.g., A£02F2" + 02~ > AJIO3" + 2F~ and A£0F~ + 02" > AJ>0~ + 2F" This produces an increase in electrical conductivity but this 44 increase is not substantial. Low electrical conductivity by the addition of A2-203 leads to higher melt rates and higher efficiency. 17 46 It has also been noted that the rate of dissolution of Aj^Og in CaF^ is higher in the presence of CaO. This is be cause A£2n3 dissolves essentially by an acid/base reaction: A£2°3 + °2~ 2AiO" and hence is dissolved much faster by a slag containing CaO than CaF^ alone. This decreases the overall melting time and thus also increases the efficiency. Due to the high temperatures attained during the melting, expecially in the region directly below the electrode, the thermal stability of the .slag components should be high i.e. the slag components should have a low vapour pressure. Also 44 it has been noted that production of CaO due to volatility of CaF^ through the reaction A£203 + 3CaF2 == 2A£F3 + 3CaO can cause significant shift in AJ^O^/CaO ratio. Presence of Si0^ and H20 will have similar effects. An essential requirement for the slag to be a suitable melting medium is that it must have a liquidus temperature at least 100°C below that of the metal, but must also form a solid phase on freezing which has a melting point higher than that of the metal.44'4'' These conditions lead to good sur face quality of the castings. 18 The phase diagram for CaF2-A£203-CaO is shown in Figure 48 11. From this it can be seen that for low liquidus tempera tures of the slag, approximately equal proportions of A&203 and CaO should be present in the ternary slag system. The metal refining characteristics are a complex function of the slag composition, power mode (a.c. or d.c), melting conditions, deoxidation, and of course the metal that is being refined. CaO is primarily used as a desulphurising agent. 44 It has been suggested that addition of a rare-earth-oxide (RE203) to the ternary system of CaF2+CaO+A&203 leads to wide regions in the phase diagram where the physical properties are compatible with ESR processing. Besides strongly complexing 2 -0 and F~ (hence increasing the electrical resistivity), the presence of (RE-0-S)n~ complexes also leads to a strongly basic behaviour with respect to sulphur reactions. Both these conditions are highly desirable for ESR. Taking the above points in consideration, two different slag compositions were selected for stainless steel and low alloy steel castings. For stainless steel we had 70% CaF2 , 15% CaO , 15% As^ and for low alloy steel (AISI 4340) where a very low sulphur content is required we had 50% CaF0 , 15% CaO , 15% A£o0o , 20% La90q 19 As the slags used were highly basic and the fact that the castings were deoxidised, careful control of residual hydrogen content in the castings was of prime importance. Three main sources of hydrogen in ESR are the hygroscopic, basic slag, the humidity of the atmosphere from which hydrogen is transported through the slag into the liquid metal pool, and the hydrogen content of the electrode. Of these the last one is not so serious as the first two because sufficiently low electrode hydrogen contents can be obtained by today's steel melting practice. Highly basic slags (e.g. slags con taining CaO) hydrate easily. Also hydrogen pick-up from the atmosphere is increased if basic slags are used because such slags promote the formation of OH" ions -02~ + H20 ^—20H" (at the slag surface) and this is subsequently transferred to the metal. 20H" ^=^-02" + [0]+ [2H](at the slag-metal interface) According to this a low partial pressure of 0^ in Fe (as in the deoxidised condition) would promote transfer of hydrogen to the metal Careful storage, preheating, prefusing, and using the molten slag start technique are a few ways to avoid hydrogen 20 in the slag. Hydrogen from the atmosphere can be reduced by creating a dry atmosphere by simply protecting the surface of the slag by a shielding hood. Application of superimposed d.c. on a.c. melting has also been claimed to reduce the hydrogen 49 50 content during electroslag remelting, ' but these were an ambiguous series of experiments. Utilizing the molten-slag start technique, hydrogen prob lems were not encountered in the castings. However, when the dry slag-start technique was used, the problem of hydrogen was severe enough to create several longitudinal holes in the bottom half of the castings. Using prefused or even double prefused slag in dry start did not solve the problem. The ternary diagram of CaF2-A&203-Ca0 (shown in Figure 11) showed that 70%CaF2/15%Ca0/15%cU203 slag composition could precipi tate some CaO when solidified and this could easily rehydrate. Therefore prefusing the slag did not eliminate hydrogen in the casting. To avoid hydrogen problems due to moisture in the slag in the dry-slag start, the components were heated to suf ficiently high temperatures. CaF2 and Aj^O^, which are not so susceptible to moisture, were preheated to 500°C while CaO was preheated and La2(C03)3 was calcined at 800°C. These were used directly in the hot condition. No traces of mois ture related porosity were observed. 21 Although it was a clear indication that the holes in the castings were due to the moisture in the slag, precaution was taken to avoid hydrogen from the atmosphere. A shielding hood made of stainless steel and lined with saffil (A^O^ > f i brous sponge) was used. This had a hole in the middle for the electrode and two more small holes at the side - one for addi tion of deoxidant and alloying additions and through the other hole, argon was passed during the melting operation. Several valve castings were made in low alloy steel (AISI 4340) and cast stainless steel (CF-8M). All the valves were subjected to non-destructive testing (NDT), macroexamination, microexamination and mechanical testing. The next section deals with the evaluation of valve cast ings and has been divided into two parts. Part A deals with the stainless steel valve castings while part B deals with low alloy steel (AISI 4340) valve castings. 22 Chapter 4 EVALUATION OF ESC VALVE CASTINGS A. STAINLESS STEEL VALVE CASTINGS Five stainless steel valves were made. Out of these two were made from CF-8M (0.08% max. C, 1.5% max. Mn, 1.5% max. Si, 0.04% max. S, 0.04% max. P, 18.0-21.0% Cr, 9.0-12.0% Ni, 2.0-3.0% Mo) electrode, one from 316 (0.08% max. C, 2.0% max. Mn, 0.03% max. S, 0.04% max. P, 1 .0% maxsi,' 10.0-14.0% Ni, 16.0-18.0% Cr, 2.0-3.0% Mo) rolled bar, one from 316 rolled bar with controlled addition of chromium chips and the last one with 316 rolled bar from controlled additions of ferro-chrome and ferro-molybdenum powder. The alloying was done to change the 316 bar chemistry to CF-8M composition. This was done keeping in view that even a manu facturer who does not have electrode casting facility can pro duce CF-8M castings with the readily available 316 rolled bar composition. Besides this, a CF-8M block (12" x 16%" x 17") was con ventionally cast after refining in A0D. The heat treatment given was a double quenching treatment from 1094°C (2000°F). The properties of this conventional casting were compared with ESC valves. 23 A typical electroslag casting of a valve body made in stainless steel is shown in Figure 12. 4.1 Remelting Log for Stainless Steel ESC Valves Table I gives the melting conditions for 5 stainless steel valves made by the ESC process. Some of the special character istics noticed during the melt or in the castings are as fol-1ows : -Valve No. 5 - Melting had to be interrupted abrubtly (hence no hot topping cycle) because of a minor explosion heard towards the end of its run. This will be discussed in section 4.3.7. To avoid any porosity due to the moisture in the slag components (particularly CaO), the slag was prefused and crushed for use. However, when the casting was sectioned, large holes (about the diameter of a pencil) running longi tudinally from the bottom for a distance of about 8 inches were observed.(see Figure 1.3). As the slag was prefused, the chances of moisture being responsible for the holes were low. They could also be due to carbon monoxide evolution due to in sufficient deoxidation. Valve No. 6 - The casting was incomplete because the electrode was not long enough. Extra aluminum powder was added in the beginning of the run to avoid CO evolution Also the mold was covered from the top and argon was passed over the slag. The ;sectioned : casting again showed 24 longitudinal holes. At this point it was certain that the holes were due to moisture. Valve No. 7 - Electrolytic chromium chips were added to increase the chromium content of the casting. The slag was prefused twice to avoid the moisture problem but again there were some holes. However, the size and density of these holes had decreased considerably. As mentioned in section 3.4 the holes were due to rehydration of precipitated CaO in the fused slag. Valve No. 9 - Heating the slag components to high temperature (mentioned before) avoided the problem of moisture in the slag. No trace of moisture related porosity was observed. The casting was sound and complete. Valve No. 10 - 70% Fe-Cr and 65% Fe-Mo powder was added to increase the chromium and molybdenum contents of the casting. The feeding rate was uneven because of an inadequate feeding arrangement which probably resulted in clusters of ferroalloy powder segregated in pockets in the casting. No porosity was observed in this casting. 4. 2 Non-Destructive Testing (NDT) As U.B.C. does not have the facility for radiography and the appropriate ultrasonic testing for heavy sections, two 25 CF-8M valves were externally evaluated - one for radiography and one for ultrasonic testing. 4.2.1 Dye Penetrant Test The stainless valves were sectioned longitudinally and a conventional dye-penetrant system was used. It showed that the internal surface was sound. 4.2.2 Ultrasonic Test Ultrasonic testing was performed on a full ESC valve and then on a sectioned half casting. The testing was done accord ing to the ASME SA-388 (ASTM A 388-71) specification for heavy section steel forgings. Test results indicate that 'no useable ultrasonic resppnse was received from the austenitic (CF-8M) stainless steel valve casting at depths of 7-3/4 inches...:' The ASME specifications also note that the 'heavy austenitic stainless steel forgings are more difficult to penetrate ultra-sonically than similar carbon or low-alloy steel forgings. The degree of attenuation normally increases with section size; and the noise level, generally or in isolated areas, may be come too great to permit detection of discrete indications. In most instances this attentuation results from inherent coarse grained microstructure of these austenitic alloys:1 Following this a three inch thick section with one machine cut surface was tested with the 1 MH,- 3/4 inch diameter transducer. The 26 penetration direction was transverse to the grain orientation. It was still questionable whether or not fine defects could be detected. Although a better penetration could have been accomplished in the direction of the grain in a three inch section, these results indicate that ultrasonic testing is not suitable for austenitic stainless steel ESC valves. 4.2.3 ..Radiography Test Radiography was performed in accordance with ASTM E 94-77 specification using a cobalt-6'0 source. No apparent defects could be detected. 4.3 Destructive Testing (DT) The conventional and ESC valve castings were sectioned longitudinally through the middle and half inch thick plates were cut out (see Figure 14) for destructive tests. 4.3.1 Macrostructures The half inch thick plates were surface ground and macro-etched vin1 an acid solution (38% HC£, 12% H2S04, 50% H20) at about 65-70°C for 1-2 hours. The macrostructures obtained are shown in Figures 15-19. The grain structure clearly shows the chilling effect of the water-cooled mold and almost 27 vertical solidification pattern. Also, the structures are essentially free of sol i di fi cati on defect s. al though some pecul i ari.t i es:'do exist. Figure 16 shows a slag inclusion at the top of the casting. This was due to an interrupted hot-topping cycle and can be corrected easily. Figure 17 shows a change of pool profile in the top right hand corner and a peculiar structure in the centre.. This is believed to be a piece of electrode which dropped during the melting operation. Figure 18 shows some big inclusions scattered in the casting. This casting was made with the addition of ferro-chrome and ferro-molybdenum powders during the melt and these inclusions are believed to be due to the ferro-alloy additions. These two eases will be identified and discussed subsequently. The macrostructure of the top quarter part of the longi tudinal se.c.tiion of the conventional CF-8M casting is given in Figure 19. Most of the area is covered by equiaxed grains though there is some columnar structure near the edge of the casting. This macrostructure also shows a banded structure between the columnar and the equiaxed zones and at an angle to the top surface. This is thought to be due to either a feeder or a riser placed at this position. The macro-etched plates were then cut up for micro-examination and other tests. 28 4.3.2 Mi crostructures The austenitic cast stainless steel designated as CF-8M by ACI closely corresponds to the wrought grade AISI type 316 stainless steel. The silicon and chromium contents of the cast alloy are higher and the nickel content lower than in the wrought alloy. The variations in chemistry may be small but they are not trivial - they provide for optimum forgeability and considerable ductility for hot and cold working on one hand and optimum castability on the other. Depending principally on composition, but also on thermal history to a lesser extent, microstructures ranging from non-magnetic (fully austenitic as in Type 316) to perceptibly magnetic (due to substantial amounts of ferrite phase present in the austenitic matrix, as in CF-8M) can be produced. Besides production of sound castings, presence of delta-ferrite improves weldability and improves the resistance to 51 hot cracking during casting operation. It also improves resistance to corrosive :media. such as• sulphurous and acetic 52 acids, to intergranular attack, and to chloride stress 53 corrosion cracking. Delta ferrite also increases the proof stress and tensile strength values by a dispersion-strengthen-54 ing effect. (this will be discussed later). This is the real reason for higher chromium content in CF-8M. From the above we see that presence of delta-ferrite in CF-8M castings is essential to obtain good corrosion 29 resistance and mechanical properties. However, the presence of delta-ferrite in austenitic stainless steels has been shown 55 to accelerate the formation of an intermetal1ic sigma phase. Sigma phase was first detected in Fe-Cr-Ni alloys and reported 56 in 1927. It has been identified as a hard,brittle, non magnetic, intermediate phase with a tetragonal crystal structure occurring in many binary and ternary alloys of theitransition 47 58 59 elements and has been well documented in the literature. ' ' The formation of sigma phase is enhanced by addition of fer-rite-stabilizers such as chromium, molybdenum and silicon. Although this phase forms most readily from ferrite, it can 6 0 also form from austenite. Sigma formation is time- and tem perature-dependent and can be described on a transformation 61 diagram by a typical 'C curve. The temperature range of sigma formation is regarded as about 600°C to 950°C and in alloys of the 18-8 type with 2 to 3 percent molybdenum and retained delta-ferrite, the maximum rate of formation occurs at about 850°C with ferrite transforming to sigma and austenite in less than one-half hour. It has been reported ' that delta-ferrite being a region of relatively high chromium content, forms sigma in a comparatively short time by a concentration of chromium through diffusion from adjacent chromium-rich ferrite to areas where sigma nucleated. The depletion of chromium in the surrounding ferrite eventually results in the formation of austenite. This is shown in Figure 23. 30 The effects of sigma on the corrosion behaviour of austeni tic stainless steels are most serious in highly oxidising en-59 vironments such as nitric acid. In general sigma phase in creases the hardness, yield strength, and tensile strength, but decreases the impact strength, both at room temperature and at elevated temperatures. Loss of low temperature ductility and corrosion resistance is particularly severe when the sigma phase forms a nearly continuous network around the austenite grains.^2 In consequence of the above discussion, it is important to identify the ferrite and sigma phases in the stainless steel valve castings and to eliminate any sigma phase present in the as-cast condition by a suitable heat treatment. The CF-8 type cast alloys are used in the solution treated con dition which involves heating at 1100 to 1150°C and quenching in water or oil. The heat treatment adopted was heating the specimens to 1120°C for 1-2 hours and quenching in water. The microstructure after heat treatment was essentially austen itic with ferrite distributed throughout thematrix in dis continuous pools. Sigma phase was eliminated. Many techniques have been cited in the literature to identify and differentiate sigma phase from other phases occur-63 6 5 ring in austenitic stainless steels. ' Two different polishing and etching procedures were used to identify the ferrite and sigma phases in ESC valves. 31 In the first one, the specimens were mechanically polished to 5 microns, then electro-polished in chromic-acetic solution (133 ml acetic acid, 7 ml water, 25 gms Cr03) at 28-30 V for 8-10 minutes. Then they were electro-etched in 10 wt.% oxalic acid solution at 6 V for 20-30 seconds. This procedure was found to outline the ferrite phase clearly and the oxalic acid etch attacks the sigma phase very rapidly. Figure 20a-b shows the microstructure of a specimen from CF-8M valve in the as-cast condition. The ferrite phase shows up in relief and it is clearly outlined and the sigma phase which is predominantly present within the ferrite phase is attacked severely. The heat treated specimens show that the sigma phase is eliminated and only ferrite is left behind. Similarly Figures 2.1 and 22 show the microstructures from 3 different valves in the as-cast and heat-treated conditions respectively. In the second procedure, the specimens were polished to 1 micron and etched in a mixed acid solution (HNO^ and HC£ in H20) several times to remove the disturbed metal. Then they were etched in Vilella's reagent (5 m£HC£, 1 gm pjicric acid, 100 nu ethanol) to outline the sigma phase. Ferrite is faintly outlined. Then the specimens were electro-etched in 10N KOH solution (56 gms KOH, 100 m£ water) at 2.5 V for a few seconds, this stains the sigma phase yellow to reddish-brown and the ferrite to bluish-grey. Austenite is not attacked. One speci men from the conventional CF-8M casting was kept at 850°C for 32 14 hours to promote sigma formation and then it was etched with the above procedure. Figure 23 shows the microstructure of this specimen. The black regions are the sigma-phase, the dark grey regions are the untransformed ferrite phase, and the light grey area is the austenite. The austenite which has been trans formed from ferrite is also seen. As the conventional casting was received in the solution-treated condition, the micro-structures given in Figure 24 shows that only ferrite and austenite are present. Figure 25 shows the microstructure of a CF-8M ESC valve No. 9 in the as-cast condition. Sigma phase within the ferrite boundaries is easily detected. Figure .26 shows the specimens from the same valve but in the solution-treated condition. Sigma phase is completely eliminated. Microhardness tests on sigma and austenite were also car ried out to confirm that what has been distinguished to be sigma phase from metallography is actually the hard and inter-metallic sigma phase. Figure 27 shows the identation of the diamond pyramid hardness indentor on sigma and austenite phases. The relative DPH values calculated from these are -sigma phase - 643.3 DPH austenite phase - 255.0 DPH The microhardness value obtained for sigma phase is some what lower than the value quoted in literature which is 750 DPH. This difference could be because the indentation taken 33 here is over a network of sigma particles while the literature value is within a large isolated sigma particle. 4.3.3 Delta Ferrite Measurements The importance of delta ferrite in austenitic stainless steel cast alloys has been discussed briefly in the last section. It becomes essential, therefore, to determine the ferrite content of the castings. The ferrite content was measured by three dif ferent instruments - a Ferrite Indicator, a ;Magne Gage, and a Quantimet. Ferrite content was also estimated from the 66 Schoefer's diagram. The Ferrite Indicator uses magnets calibrated to known ferrite numbers to 'bracket' the ferrite content of the test piece between two ferrite numbers. The ferrite number has been adopted as the accepted index which quantifies the fer rite content. Although closely related to actual percentage ferrite content, particularly at low ferrite contents, the ferrite number does not correspond precisely to percent fer rite. Measurements were done on the specimens used for micro-examination. The surface was rough polished to avoid erron eous results due to oxidation. Table II shows the ferrite numbers obtained from different locations in the castings. As the Ferrite Indicator only 'brackets' the ferrite content, a iMagne-Gage was used to find out the specific 34 ferrite content of the castings. This instrument also works magnetically and has a calibrated coil spring attached to a pointer which moves on a dial. The dial reading indicates the ferrite content of the test specimen in ferrite numbers. The specimens were polished to 320 grit to provide an even, clean and similar surfaces to all specimens. The ferrite con tents obtained from various locations in the castings are given in Table II. Table III gives the average ferrite number of each casting in the as-cast and heat-treated conditions. From this table it can be noticed that for valve numbers 5, 6 and 7, the ferrite contents in the heat-treated condition are 1 ower than in the as-cast condition. Normally, an opposite result would be expected due to the presence of sigma phase within the ferrite bounda ries in the as-cast condition. This anomaly can be explained by the thermal history of the heat-treated specimens. The specimens from these three valves were heat-treated in stainless steel bags to avoid excessive oxidation. It was found that quenching of the specimens was not fast enough as it was difficult to cut open the bags before quench ing. Hence the microstructure still showed some sigma phase. Therefore the specimens were heat treated agaiin wi thout the stainless bags and quenched in water. It seems that this particular thermal cycle resulted in lower ferrite content. It has been indicated*^ that the residual ferrite content of duplex alloys is strongly influenced by the solutioning tem perature and the subsequent cooling rate. 35 Another point which emerges from the ferrite measure ments is that the ferrite content of the topmost part of the casting is much lower than the bulk of the casting. This is probably due to the very slow cooling rate of the top part of the casting. Towards the end of the run, there is a brief hot topping cycle and after the run has been terminated the slag blanket solidifies slowly and acts as a thermal insula tor for the last part of metal to freeze. Ferrite measurements were also done by using the Quantimet. A stain etch was done using the technique de scribed in the last section. It was found that it was dif ficult to get a good accuracy in the Quantimet. The ferrite forms a net work particularly at higher ferrite contents and it is difficult to cut out the 'snow' or noise in the normal mode. The 'snow' can be reduced by using the 'ground cut' mode and lower resolution, but this mode makes it difficult to set the threshold values which in turn leads to variable results. The use of any other more sophisticated technique was not considered to be necessary in this study. Theoretical estimation of the ferrite content was also carried out using the Schoefer's diagram. 6 6 Schoefer has developed a one-line constitution diagram for cast Fe-Cr-Ni alloys with which the ferrite content of the 36 alloy can be estimated accurately knowing only the alloy com position. It was derived from Schaeffler's diagram for weld metal. The Schoefer's diagram requires the conversion of all ferrite-promoting elements into 'chromium equivalents' and all austenite promoting elements into 'nickel equivalents' through the use of coefficients representing the ferritising or austenitising power of each element. The diagram is shown in Figure 28. Table Mil also gives the calculated ferrite con tents using the Schoefer's diagram. It can be seen the ferrite numbers estimated from Schoefer's diagram are higher than those obtained by Magne Gage particularly at higher ferrite contents. 62 Tensile data generated by other workers on C.F-8M:; shows that the strength levels are directly related to the ferrite content. Comparison of the strength levels obtained on con ventional CF-8M casting showed that the ferrite content ob tained by Magne Gage is similar to that found by those workers. Also, Schoefer's diagram is applicable to as-cast rather than heat-treated condition. As mentioned earlier the residual fer rite contents are greatly affected by the solutioning tempera ture. Quenching f/runr 1 300°C results in twice as much ferrite fi 7 as quenching from 1150°C. 4.3.4 Interdendritic Mi crosegregati on It has been frequently claimed^'^9 that interdendritic microsegregation in electroslag remelted material is greatly 37 reduced from that in the equivalent section in conventional casting. In general, low melt rates and small cross sections lead to low microsegregation. As mentioned in Section 2.1.2 an optimum melt rate is needed to achieve least microsegrega ti on . Microsegregation was studied by electron microprobe analysis of samples from two CF-8M valve castings (in the as-cast condition) and the conventional CF-8M casting (in the heat-treated condition). Line scans were performed in selected areas and across the dendritic direction. Before analysis the samples were etched and the lines to be traversed were marked with microhardness indentations such that they were 5 mm apart!. After photographing the samples in the etched and marked con dition, they were repolished before the analysis. The 'segre gation ratio' (defined as the ratio of maximum composition to minimum composition) of three alloying elements (Cr, Ni, Mo) was determined in all the samples. Figures 29a, b and 30 a-c show the line scan of the microprobe along with variation of solute concentration along the line, while Table IV gives the segrega tion ratios obtained from the analyses. The results show that-a) Out of elements analysed, molybdenum is the most highly 38 segregated element followed by nickel and then chromium. b) Interdentritic microsegregation or the segregation ratio increases from the edge to the centre of the castings. c) Microsegregation is much lower in the ESC valves than in the conventional castings for an equivalent distance from the edge. This is clearly depicted from the segregation ratio of molybdenum which is a heavily segregating element. The segregation ratio, as defined here, does not represent the true interdendritic microsegregation because the alloy in consideration has a duplex structure with delta ferrite forming interdendritical1y during solidification .process. Three stages of solidification in 18-8 stainless steels have been reported.^ The first precipitation of delta ferrite is interrupted by the precipitation of austenite and at the end of solidification there is a transition back to precipitation of delta ferrite. Austenite grows into and consumes most of the dendrites which form initially as ferrite and then at the end of solidification ferrite is precipitated from the residual melt in the inter-dendritic areas. So the segregation value determined here in cludes the partioning of alloying elements between the austenite and ferrite phases and hence does not present a true picture of microsegregation due to solidification directly from the liquid. From the line scans it can be seen that an enrichment of nickel shows a corresponding depletion of chromium as one would expect in the austenite phase. In the ferrite phase an opposite effect 39 is noticed. Therefore, the segregation ratio should be deter mined from the segregation occurring within austenite or ferrite. 4.3.5 Composition Analysis and Macrosegregation Top to bottom and side to side composition analysis of the ESC valves and the conventional casting was carried out using optical emission spectrograph, Tables V-X show the average composition of different regions of the castings along with the electrode composition for comparison. It can be seen that there is a negligible composition difference between the electrode and the casting. The advantage of ESR for as-cast material is the rather small pool volume which approximately gives the same solidifica tion conditions at the centre and surface in these relatively small sections. This results in low radial macrosegregation. However, longitudinal segregation might occur because of the change in the slag composition and oxygen potential during the melting operation.^1 Generally this is also small. -Macrosegregation in the ESC valves and the conventional casting was determined by the composition analysis mentioned above. The results with respect to top to bottom and side to side composition variation of chromium, nickel, molybdenum, and carbon are given in Figures 31-36. No definite macro-segregation trend is observed in the castings and ithe 40 variation of elements is very small. The only point which can be noticed is that wherever there is a positive deviation of chromium from the mean content, there is a corresponding nega tive deviation of nickel from the mean content and vice versa. From the variation of chromium in Valve No. 7 (see Figure 33) it can be seen that composition uniformity can easily be obtained by controlled addition of chromium during the melting operation. However, a reasonable variation in composition with respect to chromium and molybdenum was observed in Valve No. 10. In this valve ferro-chrome and ferro-molybdenum powder was added during the melt (see Figure 35). The variation is due to the inadequate feeding mechanism which resulted in unstable feeding of the powders. Two magnetic vibrator feeders set at a predetermined feeding rate were used. A funnel with a long tube was used to feed the powders.from the vnbrators to the ESC mold. During the melting operation it was noticed that the funnel was small to handle the volume flow and the powder was frequently jammed in the funnel and had to be dislodged manually. This led to an uneven feeding rate and probably resulted in sudden additions of large quantities of the powders. This not only resulted in a composition varia tion but also agglomerated into clusters of powders which could not melt completely and appeared as large inclusions in pockets in the solid casting. The latter point is discussed 41 in detail later in section 4.3.6. The use of a better feeding arrangement would easily over come this problem and a sound casting with uniform composition can be produced (as in Valve No. 7 with the addition of chromium chips). It is firmly believed that converting of AISI 316 composition to ACI CF-8M composition by addition of chromium and molybdenum will not present any insuperable production prob lems. 4.3.6 Problems With Alloying During the ESC Operation As outlined above controlled addition of chromium and moly bdenum was made during the electroslag casting operation of Valve Nos. 7 and 10. Chromium chips were added in Valve No. 7 while ferro-chrome and ferro-molybdenum were added in Valve No. 10. Although a sound, solid casting with uniform composition could be made with the addition of chromium chips, the same could not be achieved with ferro-chrome and ferro-molybdenum additions. In the macrostructure of Valve No. 10 (Figure 18), large inclusions can be seen and these have been marked on the photograph for easy identification. Observing these areas under the Scanning Electron Microscope (SEM) it became obvious that they were clusters of the ferro alloy powder. The SEM photographs of areas 1 and 2 at different magnifications are shown in Figures 37.(a-d) and 38(a-d). From these it can be seen that the ferro alloy powder did not melt completely but instead 42 were loosely sintered together forming 'small pockets' in the cast material. The powder seems to have been partially molten as a dendritic structure can be observed at 2000X magnification in Figure 38(d). To identify whether it is Fe-Cr 'or Fe-Mo pow der, the areas shown were analysed with. Energy Dispersive X-ray Analyser (EDXA). The results of EDXA on the 'parent metal' and areas 1 and 2 are shown in Figure 39. Areas 1 and 2 both show very distinct peaks for Mo and Cr. Although the 'parent metal' shows both Cr and Mo peaks, they are not as pro nounced as the ones from areas 1 and 2. From this it is obvious that areas 1 and 2 are much richer in Cr and Mo than the parent metal. To determine whether the material in the 'pockets' is Fe-Cr or Fe-Mo powder, the non-metallic inclusions present in areas 1 and 2 and also in the original Fe-Cr and Fe-Mo powder were analysed in the SEM. The EDXA graphs for the inclusions present in area 1 and 2, in Fe-Cr powder and in Fe-Mo powder are shown in Figure 40. From these it can be seen that the composition of inclusions present in areas 1 and 2 is very similar to the ones present in Fe-Cr powder. The inclusions in Fe-Mo are of completely different composition. This could be expected as Fe-Cr and Fe-Mo are produced by two entirely dif ferent processes. Fe-Cr is produced generally in an electric-arc furnace with basic dolomite lining and hence the inclusions contain Ca and Mg. Fe-Mo, on the other hand is produced by 43 aluminothermic reduction process and hence the inclusion morphology and composition is completely different. This shows that the material present in the pockets basically consists of Fe-Cr powder. . To confirm this by another method, areas 1 and 2 were analysed in the electron microprobe. The results obtained for Cr and Mo are shown in Table XI. The Cr and Mo contents of the parent metal are also given for comparison. Initially, the ferro-chrome powder had about 70% Cr and the ferro-molybdenum powder had about 65% Mo. The results show that areas 1 and 2 have about 54% Cr and 13% Mo. From this it is certain that areas 1 and 2 contain much more Cr than Mo and that it is, the ferro-chrome powder which did not melt completely and formed pockets in the valve casting. One would normally expect that Fe-Mo would not go in solu tion as compared to Fe-Cr because of its higher melting point (m.p. of 65% Fe-Mo = 1900°C, m.p. of 70% Fe-Cr = 1650°C) and also because it is heavier. So the question as to why Fe-Cr powder did not melt and formed agglomerates in preference to 44 Fe-Mo can only be answered by the fact that the particle size of the Fe-Cr powder used was larger than Fe-Mo powder and also that the feeding rate was much higher for Fe-Cr than for Fe-Mo.'(30 . 5.-. gms/min . vs. 4.5 gms/min). These inclusions occur red because of the unstable feeding of the powders as described earlier in Section 4.3.5. 4.3.7 Problems With Using Cast Electrodes in ESC Operation Towards the end of the melting operation of Valve No. 5, there was a minor explosion inside the mold. The melting was immediately terminated as a leak in the water-cooled mold could have caused this effect. After removing the casting the mold surfaces were cleaned and inspected for leaks. How ever, no trace of any leaks were found. Another source of the explosion could have been a macro porosity in the elec trode. Therefore the electrode was sectioned near the tip and indeed there was a porosity probably due to secondary piping in the cast electrode (see Figure 41). When the porosity was exposed, the entrapped pressurised gases must have been released under the slag layer causing the explo sion. Although this does not present a threat to the safety of the ESR operation, however, it does cause certain in stabilities in the melting process and should be avoided. Another problem which is much more serious was encountered 45 when a cast electrode was used. After sectioning and macro-etching Valve No. 9, an unusual pool profile was noticed in the top right part and also a foreign body having a different stru cture was observed in the centre of the casting (see Figure 17). Taking the latter point into consideration, it is thought that a foreign body (possibly an electrode piece) had dropped into the liquid metal pool during the melting operation. To confirm this speculation a piece was cut from this area and repolished and etched to reveal the structure. This structure is shown in Figure 42. It can be seen from this photograph that the structure of the electrode piece is entirely different to the parent casting. The electrode piece is composed of re-crystallised polygonal grains while the 'parent casting' has the normal dendritic structure. Although the cast electrode has a coarse dendritic structure (see Figure 43), the thermal cycle which this electrode piece went through is enough to convert the dendritic structure into polygonal grains. The composition of the electrode is more or less similar to the casting and therefore chemistry alone cannot distinguish between the electrode piece and the casting. However as the electrode was cast by a conventional method, the morphology of non-metallic inclusions in the electrode piece would be much different than the parent electroslag casting but similar to the ones present in the electrode, particularly at the elect rode tip. 46 So the inclusions from three areas (electrode piece, parent casting, and the electrode tip) were examined in an optical microscope and the SEM. Figure 44 shows the inclusions from the electrode piece and the electrode tip as observed through optical microscope. All the inclusions have a composite (multi-phase) and similar morphology although the average size of the inclusions from the electrode tip is larger. The three areas mentioned above were also examined under the SEM and the composition of the inclusions was determined by EDXA. The kind of inclusions found in the electrode piece,the electrode tip and the parent casting are given in Figure 45. : . • The corresponding compositions are given in Figure 46. The fol lowing can be deduced from these figures -a) The type of inclusions observed from the electrode piece and the electrode tip are very much alike except that they are bigger in the latter. b) The shape of the inclusions observed in the parent casting is entirely different and the size is much smaller. Also they appear to have nucleated and grown in the interdendri-tic region because they intersect delta ferrite which forms interdendritical1y. c) However, the point which confirms that it is actually a piece of electrode sitting in the middle of the valve casting is the composition of these inclusions. Figure 46 47 shows that both the electrode piece and the electrode tip have inclusions which are significantly rich in silicon and manganese. In fact their overall composition is very similar. The composi tion of the inclusions from the parent casting, on the other hand, is entirely different. They are basically alumina-type inclusions; Silicon is completely absent. All these points confirm the speculation that the foreign body observed in the electroslag casting is actually a piece which has dropped from the electrode during the melting opera tion. The peculiar pool profile observed in the casting is thought to be an event which occurred as a result of the fallen electrode piece. The probable events which took place before and after the dropping of th electrode piece are shown schematically in Figure 4-7.(a-.c.). The electrode was not in the centre of the mold during the melting process and this resulted in an asymmetric poor profile as shown in Figure 47(4). Asymmetric pool profiles are common if the electrode is off-centre. This was also ob served in another valve casting in AISI 4340 steel (see Figure 59). At such a time a piece from the electrode fell off. This i s apparently qui te common with cast electrodes. The cast elect rodes invariably have a centre porosity (perhaps representing secondary piping) and this might have resulted in a piece of electrode being detached due to melting by the slag entering 48 the porosity. After this piece of electrode fell off, the high temperature zone was shifted more towards the centre (see Figure 47(b). As a result of this the liquid metal pool on the right side was suddenly cooled and the liquidus and solidus advanced rapidly. The oriented dendritic growth was cut off and the liquid pool solidified with a fine and randomly oriented den dritic structure. The pool profile was shifted to the left and it made little or no " difference to the liquidus and soli dus on the left side (see Figure 47(c)). The affected pool pro file area is shown in Figure 48. One can notice that the oriented dendritic crystals have stopped growing and there is a band of unresolved structure which represents the solidi fication front position change through the event. Above this, a fine and random dendritic structure is observed. It has been 72 pointed out by Jackson and Mitchell that there is a struc ture refinement due to the sudden increase in growth rate and changes in primary direction due to the change in the heat flow pattern. Metals with low anisotropy of growth direction (e.g. 300 series stainless steels) will renucleate in the new heat-flow direction immediately after the change in heat flow. Com position changes due to sudden growth rate occur only in the case of elements having small segregation coefficients (e.g. C,S,P and 0). However the segregation of these elements would be difficult to detect in CF-8M as they are present in very small quantities. It would be worthwhile to mention here that the presence 49 of the electrode piece inside the valve casting would not be detected by the NDT tests (e.g. Radiography and Ultrasonic test), which are generally used to qualify the valve castings for nuclear applications. In fact such an electrode piece would be removed when the valve is machined, but even if it is present it would not be detrimental to the properties of the valve. How ever if such a 'defect' occurs in a component which would be subjected to fatigue stressing (e.g. a landing gear of an air craft), it might be a potential site for the nucleation of a fatigue crack as the fatigue properties of the electrode piece and the parent casting would be very different. 4.4 Mechanical Properties 4.4.1 Tensile Testing Tensile tests were carried out on samples from the ESC valves and the conventional casting. The tests conformed to the ASTM standards (ASTM A370-376). The samples were 0.25"in gauge diameter (see Figure 49) and they were tested in the solution-treated condition (given before in section 4.3.2). Tests were carried out both in the longitudinal and the trans verse directions. The results of these tests for Valve Nos. 5, 6 and 7 are given in Table XII while those for Valve No. 9 and the conventional casting are given in Table XIII. The tensile requirements for ferritic and austenitic steel castings (ASME SA-351/ASTM A351-72 for CF-8M) and for forged or rolled alloy 50 steel valves etc. for high temperature service (ASME SA-182/ ASTM A182-74 for 316) are also given in Table XIII. The com plete specifications are given in Appendix 1. From the results given in these tables, it can be seen that all the ESC valves except for the 316 valve (Valve No. 6) satisfy the tensile requirements of ASME and ASTM code for forg-ings and castings for valve bodies and their parts used in high-temperature service. The 316 valve falls short by about 2000 psi in tensile strength. The following two points should be noted in reference to this -1) The ASTM/ASME specifications for valve bodies and parts made out of 316 stainless steel are for rolled or forged condition. Hence the specified strengths are for a fine sized and worked material and not for a cast material (low in delta-ferrite content). 2) The cast counterpart of the rolled or forged 316 is CF-8M where the composition has been altered to give a large amount of delta-ferrite in order to ihcrease the strength values and thus compensate for the loss in strength due to the non-worked condition. These two points adequately explain the lower tensile strength observed in the 316 ESC valve. So it can be con cluded that tensile properties of the ESC valves 1 iewell within the ASME/ASTM specification provided the composition require ments are met. This is also evident if we consider the 51 properties of Valve No. 7 where the chromium content was in creased from the 316 specification to CF-8M specification (i.e. the delta ferrite content was increased). The tensile properties of the conventional casting are also within the specifications. If the results of Valve No. 9 (CF-8M) and the conventional CF-8M casting are compared, it can be seen that on average the tensile strength of the latter is higher by about 20QO psi while the yield strength is higher by about 4000 psi. This can be due to two things. Firstly the delta-ferrite content of the conventional casting is higher than ESC valve No. 9 by about 24% and secondly the grain size of the conventional casting is smaller. The latter point ac counts for the higher yield strength. On the other hand the ductility (as demonstrated by percent elongation and percent reduction in area) of the conventional casting is signifi cantly lower than the ESC Valve No. 9 and although part of this can be due to the difference in delta ferrite content, the rest can only be explained by the fact that the electro slag casting is much cleaner (w.r.t. non-metallic inclusions), has less microsegregation and has a better solidification structure. This can also be seen by considering the percent reduction in area which is more sensitive to non-metallic in clusion content. Therensa sharp decrease in percent reduc tion in area from the edge to the centre of the conventional casting while no such decrease is observed in the ESC Valve No. 9. 52 From the tensile results it also became evident that the delta-ferrite content of the castings greatly influences the 73 proof stress and tensile strength values. Irvine et al. con clude that delta-ferrite increases the proof stress and ten sile strength by a dispersion strengthening effect. The fer rite has a higher yield stress than the austenite and the strain concentration in the softer austenite phase causes it to work harden to a strain greater than the nominal 0.2%, and therefore gives a higher proof stress value. In the case of the tensile strength, about 80% of the strengthening due to delta-ferrite increases is due to partitiQhing of carbon and nitrogen to the austenite, thereby increasing the work hardening rate. 6 2 Beck, Schoefer et al . concl uded from the resul ts from about 62 CF-8M and 277 CF-8 heats that the tensile and yield strengths are directly related to the ferrite content. Figure 50 shows the variation of longitudinal tensile and yield strengths with ferrite number for the ESC valves. It can be seen that the strength is a direct function of the ferrite content. However, ductility (as deter mined by percent elongation) does not show any definite relation to the ferrite content. When the average tensile and yield strengths of the conventional casting are plotted on the graph, the former lies about 1000 psi below the line while the latter show a positive deviation of about 4000 psi from the line. This higher value of the yield strength is quite significant and could be explained by the fact that the conventional casting has a much smaller grain size than the 53 electroslag casting and this results in a higher yield stress. This is also obvious from the fact that the yield strength of the samples from the edge of the conventional casting is about : . 2-1.5% higher than the samples away from the edge. The fractured and deformed areas of the 0.25 inch diameter tensile samples exhibited the characteristics of a large grain sized material (see Figure 51). Coarse grain materials approach 64 more closely the behaviour of single crystals. To present a valid result, a sufficient number of grains in the cross section of the specimen are required so that the relatively free grains near the surface are too few in number to affect appreciably the properties of the whole aggregate of grains. In the case of tin, Pel 1-Walpole74 found that the tensile strength of rolled material increases by 60-100% with increase in the number of grains for 1 to 20-30, and very slightly increased by further refinement. He also concluded that it is the number of grains in the cross-section rather than the absolute grain size which controls the grain size/ultimate tensile strength relationship. Work done on polycrystal 1 ine aluminum also shows that the coarse grained specimens work harden substantially less than the finer grained speciments, although the general form of the stress-strain curves is similar, and the effect of grain size on the 75 elastic limit is particularly marked. It became obvious, particularly from the 316 tensile 54 samples, that the grain size of the ESC valves was large and there were very few grains in the cross-section of 0.25 inch diameter samples. As a result the tensile properties of the small specimens did not represent the bulk tensile properties. To overcome this problem, it was decided to test larger speci mens.. Large size specimens with a -gauge diameter of 1 inch and a gauge length of 4 inches (see Figure 49)„were machined from Valve No. 6 (316) and Valve No. 9 (CF-8M). These were solution treated (in the same way as mentioned before) prior to the final machining operation and were tested in the MTS machine. These specimens were 16 times larger in the cross-sectional area than the smaller specimen and therefore should have about 16 times more grains in the cross section. The results of these specimens are given in Table XIV. The de formed region near the fractured surface still shows the pre sence of large grains and the slip lines on the grains are clearly visible. The fractured surface shows an irregular out line. A close up of one of the broken tensile samples is shown in Figure 52. To measure the projected area of the fractured surface, both the ends of the.sample were photo graphed and then the area was measured by a planimeter. The average value of both the ends was taken as. the fractured area for that sample. Comparison of the tensile results of the large and small specimens (see Tables XII, XIII and XIV) shows that the yield 55 as well as the tensile strength of the larger specimens were lower than the smaller specimens for both 316 and CF-8M. The percent elongation of the larger specimens was similar to the smaller specimens in the case "of' 316 and much lower in the case of CF-8M, while the reduction in area for CF-8M was similar for both sized specimens. This is completely opposite to what was expected. It is possible that because of the difference in the size of the samples, there was a variation in the heat treat ment which could have occurred during the heating or quenching of the specimens, and that this had a much greater influence in the mechanical properties than the number of grains in the cross section. So to investigate this possibility small samples were machined from the undeformed region in the threaded grip section of the large tensile specimens from Valve No. 9 (CF-8M). In all three tensile samples (two from the edge and one from the centre) were tested. The results (given in Table XIV) show that the yield and ultimate tensile strength and percent reduc tion in area are- only marginally higher than the large samples This shows that the difference in heat treatment of the small and large specimen is responsible for the difference in mechani cal properties rather than the number of grains in the cross section. In conclusion it can be said that the tensile properties of the ESC valves can easily meet the required ASME/ASTM codes if the chemical composition is controlled within the required range. 56 B. LOW ALLOY STEEL VALVE CASTINGS Six valves were made using AISI 4340 (0.38-0.43% C, 0.6-0.8% Mn, 0.04% P, 0.04% S, 0.20-0.35% Si, 1.65-2.00% Ni , 0.70-0.90% Cr and 0.20-0.30% Mo) rolled bar as electrode. 4.5 Remelting Log for AISI 4340 ESC Valves Table XV gives the melting conditions for the ESC valves. Towards the end of the melting in Valve No. 3, the cables con nected to the top mold were heated up and started smoking. This was because the casting lost contact with the base plate and so most of the current was carried by the cables. These cables were not designed to carry such high currents and so they were heated. The inside of the mold (where the cables were connected) was not damaged. In all further castings, this problem was overcome by joining the base plate and starter pi ate with screws. The slag composition was changed after Valve No. 3. A rare earth oxide {La^Q^) was used. Valve No. 8 was made without any problems and the casting was sound. However, Valve Nos. 11 and 12 had to be rejected because of porosity due to moisture. Only dimensional measurements could be done on these valves. Valve Nos. 13 and 14 were also sound. A typical electroslag casting of a valve body in AISI 4340 steel is shown in Figure 53. 57 4.6 Non-Destructive Testing (NPT) 4.6.1 Dye Penetrant Test As Valve Nos. 11 and 12 showed porosity due to moisture, the other valves were sectioned and tested for cracks. Valve Nos. 3, 8 and 13 did not show any crack. However, Valve No.;. 14 showed cracks all over the section (see Figure 54). This valve was heat treated as one whole piece and these cracks were due to improper heat treatment. This point will be dis cussed later in this chapter. 4.6.2 Ultrasonic,; Test Unlike CF-8M valve castings, AISI 4340 valve castings did not present any problems in ultrasonic testing. The full section thickness could be penetrated. The results of ultra sonic testing on two castings showed that they were sound. 4.6.3 Radiography Test Radiography was performed using the cobalt-60 source. The radiographs obtained were satisfactory and no flaws could be detected. The radiographs of Valve No. 3 are shown in Figure 55. 4.7 Destructive Testing (D.T.) The sectioning procedure has been mentioned in Section 4.3. 58 4.7.1 Macrostructures The AISI 4340 steel was macroetched in an acid solution ()50% HC£, 50% H20) at about 65-70°C for 1-2 hours. The macro-structures of longitudinal sections from Valve Nos. 3, 8, 13 and 14 are shown in Figures 56-59. One-half of the section from Valve No. 3 was etched in 10% animonium persulphate solu tion. Figure 60 shows a transverse section through a mild steel ESC valve. All the macrostructures are free of defects except the one from Valve No. 14 as shown in Figure 59. Cracks running longitudinally (probably inter-dendritic) can be seen. The top part (which was etched for a longer time) clearly shows the cracks. Also a slag entrapment on the top indicates in sufficient hot-topping cycle. The macrostructure also shows that the pool profile was assymmetric and this must be due to the electrode being off-centre with respect to the mold. Some indications of this are also noticed from the macrostructure of Valve No. 8 (shown in Figure 57). From the macrostructure of the transverse section it can be seen that the grain size near the edge (surface of the casting) is smaller than in centre. Specimens from edge, mid-radius,;centre and top of Valve No. 3 were etched in Oberhoffer's solution (30 gms FeC^* 2 gm CuC£2, 0.5 gms SnC£2,350 nu HC£, 500 nu ethyl alcohol, 500 nu distilled water).to reveal the dendritic structure. These are shown in Figure 61(and). The dendrites are fine at the edge and become coarser towards the centre, but they are all 59 oriented directionally. However, the photograph of the speci men from the top shows coarse dendrites in random orientation forming equiaxed grains (due to very slow cooling at the top.) 4.7.2 Sulphur Prints Sulphur printing reveals the distribution of sulphur in ingots as well as rolled or forged products. Sulphur prints were taken from Valve Nos. 3 and 8 along with a piece of electrode. These are shown in Figures 62 and 63. The valves do not show any sulphur segregation and the prints are lighter than those from the electrode indicating that the sulphur con tent of the valve castings is much lower than that of the electrode. 4.7.3 Interdendritic Microsegregation Using the same technique as described in Section 4.3.4, microsegregation at the centre of the AISI 4340 ESC valves was determined. The line scans of the microprobe along with variation of solute concentration of Cr, Ni, and Mo are shown in Figures 64-66. Also Table XVI gives the segregation ratios obtained from the analyses. From this we see that the segrega tion ratio is maximum for Mo. Also, the segregation ratio of Mo in Valve No. 13 is low while no value could be given for Valve No. 14 as the minimum count was similar to the background count. The low segregation ratio of Mo in Valve No. 13 could 60 be due to the heat-treatment and/or the fact that the specimen was located away from the centre due to the machined hole. 4.7.4 C om p o s-i-ti on Ana 1 y si s and Macrosegregati on Composition variation of chromium, nickel, molybdenum and carbon was studied in the AISI 4340 ESC valves. The variation of these elements along the height and width of the castings are shown in Figures 67-70, while Tables XVII to XX give the complete composition analyses of the starting electrode and the castings. No appreciable change in chemistry due to remelting is observed except in the case of sulphur. Besides using optical emission spectroscopy, the sulphur con tent of the valve castings and the electrode was also deter mined by "combustion analysis" using Leco Sulphur Analyser. The results are shown in Table XXI. From this we can see that the sulphur content is decreased by about 70% through electroslag processing of the material, even when the sulphur content of the starting material was low. 4.7.5 Heat Treatment and Microstructures The mechanical properties of steel are greatly affected by the heat treatment and hence the resulting microstructure. All the mechanical tests were done in the heat-treated con dition. Unmachined individual specimens from Valve Nos. 3 61 and 8 were austenitised at about 845°C and oil quenched. They were tempered at about 480°C and 550°C. The exact heat treat ment done for each set of tests are given in the mechanical property section. Valve No. 13 was machined with a vertical hole of 3 inch diameter and a horizontal hole (through the side protrusions) of 2 inch diameter. The machined valve is shown in Figure 71(a), while Figure 71(b) shows a longitudinal section of this valve. Valve No. 14 was not machined. Instead of heat treat ing individual specimens (for mechanical testing) from Valve Nos. 13 and 14, both these valves were heat treated as a whole prior to the sectioning step. Both these valves were austeni-tised in a salt bath at 845°C and oil quenched. They were tempered at 560°C. After the heat treatment, the valves were sectioned for the destructive tests. When Valve No. 14 was cut longitudinally, very fine cracks were observed. These cracks showed up clearly when a dye pene trant test was done (see Figure 54) and also when the plates were macroetched (see Figure 59), Macroetching also revealed that the cracks were interdendritic and they were concentrated inside the casting rather than on the surface. A bar with 0.5 inch square cross section cut from this valve separated along a crack without much force. The separated surfaces were smooth and shiny and the dendritic structure was revealed (see Figure 72). The machined casting did not show any such cracks. 62 From the above results it can be said conclusively that the cracks in Valve No. 14 were caused due to improper heat treat ment. Hardness measurements were done to find out the varia tion of hardness along the height and width of the heat-treated valves. These are shown in Figures 73 and 74. Valve No. 14 shows a large variation in hardness, particularly along the width of the casting. The surface shows a higher hardness than the centre. In Valve No. 13 the hardness variation is not as large. This is due to the difference in cross section between the two valves. Valve No. 14 had a section thickness of about 8 inches compared to 2% inches in Valve No. 13 (due to the machined hole). The above points strongly suggest that valve castings should be rough machined to their final shape prior to heat treatment. Beside these, other advantages include reduced heat treating cycle due to smaller cross section and weight, and that it would be easier to machine the valve in the as-cast condition because it is softer than in the heat-treated con dition. For microstructural examination, the specimens were polished and etched in 2% nital. The microstructures obtained in as-cast and heat-treated conditions are shown in Figure 75(a-d). 63 The as-cast condition has a coarse bainitic structure while the quenched and tempered condition shows a structure composed entirely of tempered martensite. Same is true for the quenched and tempered specimen from the electrode. The microstructure of a specimen from Valve No. 13 showed a different structure.Figure 76.shows that the microstructure in the interdendritic area is different than that in the dendrites. At higher magnifica tion it is seen that the light areas (dendrites) have a bainitic structure while the darker areas shows the presence of marten-site. Some ferrite is also present. This mixed microstructure resulted is low ductility when tensile test was done on this material (see Table XXIII) - This microstructure is largely due to improper heat treatment, but interdendritic segregation has also contributed to it. This suggests that 'normalising' might become an important step in heat treatment process. Heat treat ment of AISI 4340 steel (homogenising effects on mechanical properties, overheating during normalising, etc.) is well docu mented in the literature, and we believe that heat treatment of ESC valve would not present any problems. 4.7.6 Dimensional Measurements Dimensional measurements were done on the mold and valve casting Nos. 11, 12 and 13. After the mold has been assembled, the inner dimensions between certain points were taken. After the casting was made, measurements between the same points were again taken. Also the slag thickness at these points was 64 determined. From these data, 'real percent shrinkage' and 'pseudo percent shrinkage' were obtained. These are defined below and are listed in Table XXII. real percent shrinkage Di - Df - 2Ts 100 Di pseudo percent shrinkage =. Di - Df 100 L 1 where Di - inner dimension of the mold Df - final dimension of the casting Ts-slagthickness. The 'pseudo percent shrinkage', as determined here, is of practical significance and importance to foundrymen. The maxi mum value for this is around 4.6%. It was found that the maxi mum 'shrinkage' is at the corners and edges of the castings. This is because of the very high heat transfer and hence large slag thickness at these points. 4.8 Mechanical Properties 4.8.1 Tensile Testing As mentioned in Section 4.4, tensile tests were carried out on AISI 4340 ESC Valve Nos. 3, 8 and 13. Both the longi tudinal and transverse samples were tested in the quenched and tempered condition. The details of the heat treatment have 65 been mentioned before. The results are given in Table XXIII. Tests were also done on AISI 4340 rolled bar and the results are given in Table XXIV. Some tensile results which have been quoted in literature for air melted, vacuum are remelted and electroslag remelted AISI 4340 have also been included in Tables XXV to XXVII for comparison. The yield and tensile strengths are not so important as different strength levels can be achieved by suitable heat treatments. However, ductility at a certain strength level is important, particularly for transverse specimens. In ESC valves, the ductility (as repre sented by percent reduction in area and percent elongation) of the longitudinal specimens is more or less similar to that of the transverse specimens. However if we consider the conven tional rolled bar, the longitudinal values of % RA and % Elonga tion are 150% and 90% higher respectively than the transverse values at the same strength level. Obviously the bar had been subjected to extensive plastic deformation and hence the ductility is ani sotropic. ' .:. The mechanical properties (particularly ductility) can drop drastically if AISI 4340 ,steel is subjected to improper heat treatment. When the tensile specimens from Valve No. 13 were tested, they exhibited very poor ductility values even at much lower strength (Table XXIII). » .'The drop is directly related to the microstructure. As mentioned earlier, Valve No. 13 was heat treated as one piece and due to inadequate heat-treatment it showed the presence of bainite, tempered martensite and 66 some ferrite (see Figure 76). When individual tensile speci mens were re-heat treated (austenitised, quenched and then tem pered) the tensile properties were recovered because the micro-structure showed that only tempered martensite was present. Considering the tensile properties of the AISI 4340 ESC valves in general, and comparing them to the longitudinal ten sile properties of hot-rolled ESR 4340 steel (given in Table XXVI) and to the transverse tensile properties of forged ESR 4340 steel (given in Table XXVII), it can be seen that the ESC specimens show slightly lower ductility values than the longi tudinal specimens but similar ductility values as the trans verse specimens. It is noteworthy that the transverse tensile properties given in Table XXVII for forged AISI 4340 steel are perhaps the best one can obtain. For the same reasons as stated in Section 4.4, larger tensile specimens of 0.75 inch gauge diameter (see Figure 49) and therefore having 9 times larger cross-sectional area than the small specimens, were also tested in the NITS machine. The optical fractographs of these large specimens are shown in Figure 77. The results (given in Table XXVIII(a)) do not show any improvement in tensile properties over the small specimens. To eliminate any possible effect of heat-treatment, orienta tion etc. on the mechanical properties, small specimens were cut from the threaded region of the large tensile specimens. 67 The results are given in Table XXVII(b) , and they are similar to the rest. This shows that number of grains in the cross-section does not have significant effect on the tensile pro perties. 4.8.2 Impact Testing Standard Charpy V-Notch impact tests conforming to ASTM E23-72 specifications were carried out on ESC Valve Nos. 3, 8 and 13 and also on AISI 4340 rolled bar' electrode. The tests were done in the quenched and tempered condition. Specimens from both longitudinal and transverse directions were tested in Valve Nos. 3 and 13 and in the AISI 4340 rolled bar elect rode while in Valve No. 8 specimens from longitudinal, trans verse (having a longitudinal and transverse notch) and from the edge were tested. Figure 78 shows the specimen and notch orientation of the charpy specimens. The results of all these tests are presented in Figures 79-81. Each point represents an average of at least 3 tests. It was difficult to estimate the percent brittle fracture from the broken charpy specimens as the fracture surface was very irregular. Hence it was also difficult to plot percent brittle fracture versus temperature and thus determine the Fracture Appearance Transition Tempera ture (FATT). From Figure 79 which shows the charpy results of Valve No. 3 and the electrode, it can be seen that at the same hardness 68 level, the longitudinal specimen from the electrode exhibit better toughness values than the longitudinal specimens from the ESC valve. However, the transverse values of the elect rode are much lower than the ESC valve. Also the transition temperature for the longitudinal specimens from the electrode seems to be higher than for the longitudinal specimens from the ESC valve. Figure 80 shows the Charpy results of Valve No. 8. The longitudinal specimens seem to have higher energy values than the transverse specimens with a longitudinal notch. Although the transverse specimens with transverse notch (TT) show better toughness than the transverse specimens with longi tudinal notch (TL), nothing conclusive can be said as this difference could entirely be due to the difference in hard ness. However, the ductile-brittle transition character istics of the specimens from the edge are much superior as it shows reasonable toughness at very low temperatures. This could again be due to the small grain size at the edge. The ductile-brittle transition characteristics of Valve No. 13 are shown in Figure 81 and they are similar to Valve No. 8, but the curves seem to have shifted to the right. Table XXIX gives the FATT valves estimated from the 69 ductile-brittle-transition characteristics. They show that the FATT is lower for the ESC material. The optical fractographs of the specimens from ESC Valve No. 8 with various notch orientations are shown in Figures 82(a-d) and the longitudinal and transverse specimens from the electrode are shown in Figure 83. There is a great difference in fractography between the ESC valve and the electrode speci mens. The longitudinal specimens from the electrode show an even, smooth fractured surface with large shear lips character istic of high energy fracture. The transverse samples from the electrode, on the other hand show an irregular"laminated structure and shear 1','i-ps are absent even at 100°C. The speci mens from the ESC valve, on the other hand, exhibit entirely different fracture characteristics. The fracture surface is very irregular and there are areas where vertical ridges are present and the surface is smooth and shiny. This is clearly shown in Figure 84(a). These ridges are predominantly present at low temperatures. To distinguish the mode of failure in different regions of the same specimen, the specimens were examined under the SEM. The general appearance of a broken charpy specimen is shown in Figure 84(a). Areas where 'cracks' or very small ridges appear are shown in Figure 84(b). The cleavage fracture is quite apparent. Four different areas with different fracture 70 characteristics were identified. Instead of characterising them as ductile/brittle fractures, they have been termed as 'low', 'intermediate' or 'high energy' fractures. The dif ferent areas are shown in Figures 85(a-d) and identified below:-Area 1 - This is the region on the vertical sides of the ridge area. This represents 'low energy fracture' as flat and faceted fracture mode is seen. There are also some round areas where tearing is observed (probably large dimples). Area 2 - This is the region at the base of the ridge area. The fractograph..from this area shows very coarse dimples with some flat areas. This can be termed as the 'transition zone' with 'intermediate energy fracture'. Area 3 - This represents the general area (all except the ridge and the shear lip areas) and shows that fracture is due to microvoid-coalescence but here also the dimples are still coarse. This is the 'high energy fracture' area. Area 4 - This comprises of the shear lips area on the i specimen. The fracture is very much like Area 3, but here the dimples are much finer. This is also the 'high energy fracture' area. These areas are representative of all the charpy speci mens from ESC valves. This kind of fracture is probably due to firstly the large grain size and secondly due to the cast structure of the ESC material. Chapter 5 71 OTHER TRIALS AND FUTURE WORK Although the present work shows that the electroslag cast ing of valve bodies is a viable process, some more work is needed in two major areas which might lead to lower costs and improved capabilities of the process. The first area of interest is the manufacture of hollow electroslag cast valves (i.e. valves with the centre hole). Much of the work done on this aspect of ESC is again of Soviet origin and Paton and other workers have shown the feasibility 3-5 77 of the process. ' Figure 86 shows the hollow ESC valve made in the Paton Institute, U.S.S.R., and sent here for evalua tion. The two methods used by the Soviets are the fixed mandrel with the valve being cast in an inverted position and the moving mandrel with the valve in the upright position. These are shown in Figure 87. The problems associated with casting of hollow ingots (not valves) with different techniques have been dis-78 79 80 cussed to some extent by Bhat ' and Hoyle, and these might provide some useful information as to which of the two methods mentioned above would be better. A comparative study of pro perties of hollow and solid ESC ingots of the same size has been 81 done by Paton et al., and their results show that the hollow ingots have a better structure and mechanical properties. Also, hollow valves would lower the machining cost. Comparative 72 studies should therefore be conducted on hollow and solid ESC valves taking into account the economics of both the process routes. The other area where investigations are required is the technique whereby separately prepared auxiliary parts are joined or fusion welded to the main body during the casting process. By using this technique it is possible to join inlet/ outlet flanges (which have been made by the ESC process or otherwise) to the main valve body. Hence more complicated shapes which are close to the final product could be made. The Soviets have used this technique to produce very large crankshaft (with a pin diameter of 480 mm.) for diesel engines used in ships.** Some preliminary work was done here to study this aspect of the ESC process. A flange mold was used to make flange castings joined to a thick plate. This is shown in Figure 88. Then the segment of the valve mold which had the cupped side was replaced with a similar segment but with a hole in the side. An electrode piece was placed in this hole to act as an insert and the valve casting was made. This casting is shown in Figure 89. The macrostructure of the longitudinal section of this casting is shown in Figure 90. From this it' can be seen that the fusion of the insert to the main valve 73 body is possible. The incomplete fusion at the top can pro bably be corrected. A flange casting could therefore be placed instead of the insert and hence joined to the main valve. More work is needed in this area and the soundness and mechanical properties of the weld region should be studied. Both these techniques warrant further investigation and if they appear to be acceptable from the point of view of pro perties and economics, they would certainly widen the scope of the ESC process and make it more versatile. 74 Chapter 6 SUMMARY AND CONCLUSIONS The results reported above clearly indicate that the electroslag casting process is a viable technique to produce valve bodies. Non-destructive tests show that although ultra sonic testing can be used in the case of low alloy steel ESC valves, the same cannot be applied to austenitic stainless steel valves due to their inherent large grain size. Radio graphy would thus become an important tool to qualify the stainless steel castings. The macrostructures show that the castings are essenti-; ally free of solidification defects. However special pre cautions are required firstly to avoid moisture related poro sity in the castings and secondly when cast electrodes are used. Heat treatment also becomes an important step to achieve the desired microstructure resulting in optimum mechani cal properties. Chemical composition of stainless steel cast ings greatly affects the ferrite content, the presence of which is essential to obtain good mechanical properties and corrosion resistance. Although the interdendritic microsegregation increases 75 from the edge to the centre of the valve castings, it is re duced from that in the equivalent section in conventional casting. This is particularly true for Mo in stainless steel castings. It is noteworthy that the CF-8M valve casting was tested in the as-cast condition while the conventional CF-8M casting was tested in solution treated condition. The segrega tion ratio in stainless steel castings, however, should be determined from the segregation occuring within austenite or ferrite phases. Macrosegregation studies show that the variation of elements (Cr, Ni, Mo, C) along the height and width of the valve castings is small. Composition uniformity can be achieved by controlled addition of alloying elements during the melting operation with a proper feeding arrangement. The tensile properties of. the stai nl ess steel valves can easily meet the required ASME/ASTM codes if chemical composi tion is controlled within the required range. Although the strength values of the ESC valves were slightly lower than that of conventional CF-8M casting., the ductility values were significantly higher. Strength of these valves was found to be a direct function of the ferrite content. At about the same strength level, the longitudinal and trans verse ductility of AISI 4340 ESC valves was significantly 76 higher than the transverse ductility of the conventional rol led bar and equivalent to the transverse ductility of the forged ESR steel. Increasing the cross-sectional area of the tensile specimens did not alter the measured properties show ing that the number of grains in the cross-section does not have a significant effect on the measurement of tensile pro perties in these samples. The impact results show that the ductile-brittle-transition characteristics of the ESC material are better than the conventional rolled material. The impact strength of the ESC material was much higher than the trans verse impact strength of the rolled material. Also the estimated FATT was lower for the ESC material. The fractured surface of the ESC material was very irregular due to large grain size and as-cast structure. Electroslag casting of hollow valves and joining of individually cast components to the main valve body during the casting operation are promising potential applications of the process and need further investigation. At this point two problems need attention: The problem of code qualification for the ESC material must be resolved. Although it is a casting, it has been shown to be of a high quality and therefore should be permitted to be used in applications where codes presently require forgings. Also, 77 if fabrication of the final shape is done with simultaneous casting and welding, the codes should accommodate this. However, the real problem does not lie in the product but in the process control. The question is - can the ESC process be controlled such that the electroslag casting can be qualified by a generic test on the electrode stock rather than on the individual castings? This might be difficult if alloying additions are made during the casting operation be cause there will be appreciable differences between the elect rode and the ESC analyses. However, if such a code qualifica tion cannot be established, the process might become economi cally less feasible due to high testing costs. In conclusion, we may say that electroslag casting pro cess can be used to cast simple valve-body shapes in stainless and low alloy steel from cast or wrought electrodes. The properties of the ESC valves can easily meet the required specifications with proper heat treatment. The ESC valves are therefore equivalent to or better than commercial castings and forgings and can be substituted for them without any penalty. 78 REFERENCES 1. A. Mitchell: •'Modern Casting', Nov. 197.8, p.86. 2. A. Mitchell and A. Akhtar: ASME MPC-6, 'Effects of Melting and Processing Variables on the Mechanical Properties of Steel: ed. , G.V. Smith, 1 977 , p.l. 3. V. I. Rabinovich et al.: 'Special Electrometallurgy', 1972, Kiev. p.231. 4. B.E. Paton et al . : ibid, p.169. 5. B.E..Paton et al.: Proc. 3rd International Symposium on ESR, Pittsburgh, 1971, p.135. 6. B.E. Paton et al. : Report of E.O. Paton Electric Welding Institute, Kiev, U.S.S.R. 7. B.E. Paton et al.: Proc. 5th International Symposium on ESR, Pittsburgh, eds. G.K. Bhat and A. Simkovitch, 1974, p.239. 8. B.E. Paton et al.: Proc. 4th International Symposium on ESR, Japan, June 1973, p.209. 9. E.F. Dubrovskaya et al.: 'Vest Mashinost !, 1975, IV, p.62 (cited as Ref. 17 in Ref.2). 10. A. Ujiie et al . : 'Electroslag Refining', Iron and Steel Institute, London, 1973, p.113. 11. A. Ujiie et al . : Proc. 4th International Symposium on ESR, Japan, June 1943, p.168. 12. A. Ujiie et al. : Proc. 5th International Symposium on ESR, Pittsburgh, eds. G.K. Bhat and A. Simkovitch, 1974, p.251 13. M. P. Braun et al. : 'Electroslag Casting', Kiev, Na kova Dumka, 1 976 , p.44 (cited as Ref. 21 in Ref. 2). 14. B.E. Paton et al'. : 'Fonderie', 1 974, vol. 29, no. 340, p.435. 15. Demidov et al.: Issled Protsessov Obrabch Metall Davleniem, 1969, vol. Ill, p.173 (cited as Ref. 23 in Ref. 2). 79 16. G.A. Boiko et al: !'Rafinivuynshchie Pereplary', 1974, vol. I, p. 1 38. 17. I. Petram: Proc. 3rd International Symposium on ESR, Pittsburgh, 1971, p.108. 18. H.J. Wagner and K. BarAvi: 'Metals Technology', Nov. 1979, p.420. 19. A. Mitchell et al . : Conf. Proc, Institute of Mechanical Engineers, London, 1980, p.87. 20. A. Ujiie et al.: U.S. Pats. #3 ,683 ,997: #3 ,878 ,882: #3,892,271: #3,894,574. 21. B.I. Medovar et al.: U.S. Pats. #3,896,878: #3,878,882: #3,892,271: #3,894,574. 22. D.M. Longbottom: U.S. Pats. #3, 902,543. 23. A. Schneidholfer: U.S. Pats. #3,804,148. 24. W.E. Duckworth and G. Hoy1e : 'Electros1ag Refining', Chapman and Hall Ltd., 1969. 25. A Mitchell: 'Electroslag and Vacuum Arc Remelting Processes', to be published in Electric Furnace Steel-making, AIME publication. 26. National Materials Advisory Board: 'Electroslag Remelt ing and Plasma Arc Remelting', NMAB Publ. , 324, National Academy of Sciences, Wash. D.C. 1975. 27. R.H. Nafziger and others: 'The Electroslag Remelting Process', Bulletin 669, U.S. Bureau of Mines, 1976. 28. B.I. Medovar et al.: 'Electroslag Remelting1 JPRS Report 2221 7 , JPRS, Wash. D.C, 1 963. 29. U. Yu. Latash et al.: 'Electroslag Remelting', NTIS Translation AD 730371, U.S. Dept. of Commerce, Spring field, Virginia, 1971. 30. D.A.R. Kay: 'Special Electrometallurgy', 1972, Kiev, p. 63. 31. A. Mitchell: 'Journal of Vacuum Science and Technology', vol. 7, no. 6, p.563. 80 32. A.D. Wilson: ASM Technical Report System No. 76-02. 33. A.D. Wilson: Proc. Conf. TMS-AIME Ferrous Met. Committee, 106th Annual Meeting, Georgia, March 1977. 34. R.S. Cremisio et al.: Proc. 3rd International Symposium on ESR, Pittsburgh, 1971. 35. R.H. Elwell et al.: ASME-MPC-6, 'Effects of Melting and Processing Variables on the Mechanical Properties of Steel', ed. G.V. Smith, 1 977 , p.41. 36. V.L. Myzetsky et al.: 'Special Electrometallurgy', 1972, Ki ew, p. 11 9 . 37. A. Mitchell: Ironmaking and Steelmaking (Quarterly), 1974, no. 3, p.172. 38. W. Holzgruber: Proc. 5th International Symposium on ESR, Pittsburgh, eds. G.K. Bhat and A. Simkovitch, 1974, p.70. 39. Westinghouse Electric Corporation Report on 'The Qualifi cation of an ESC Valve' 1980. 40. L.M. Jose: U.B.C. Met. 499 Project Report, 1980. 41. G. Sidla and A. Mitchell: 'The Design, Construction and Operation of an ESC Installation', Special Report to DREP/DSS, June 1980. 42. R.S. Cremisio and E.D. Zak: Proc. 4th International Symposium on ESR, Japan, June 1973, p.137. 43. A. Mitchell et al: Paper presented at Warwick Conference, England, Sept. 1980. 44. A. Mitchell: 'Canadian Metallurgical Quarterly', vol. 20, no. 1, 1981 , p.101. 45. A. Mitchell and J. Cameron: 'Metallurgical Transactions', vol. 2, Dec. 1971 , p. 3361 . 46. A. Mitchell and R.M. Smailer: 'International Metals Reviews', nos. 5 and 6, 1979, p.231. 47. A. Mitchell and M. Etienne: Trans, of the Metallurgical Society of AIME', vol. 242, July 1968, p.1462. 48. K.C. Mills and B.J. Keene: 1 I nternat i on a 1 Met:a.Is Revi ews , no. 1 , 1981 , p.21. 49. W. Hotzgruber et al.: 'Special Electrometallurgy', 1972, Kiev, p.161. 81 50. H. Jaeger et al.: Proc. 5th International Symposium on ESR, Pittsburgh, eds. G.K. Bhat and A. Simkovitch, 1974, p.306. 51. R.F. Steigerwald: 'Corrosion', vol. 33, no. 9, Sept. 1977, p.338. 52. ASM Metals Handbook, 8th Edition, vol . 1, p.433. 53. J.W. Flowers et al.: 'Corrosion', vol. 19, 1963, p.!86t. 54. F.B. Pickering: 'The Metallurgical Evolution of Stain less Steels', ASM Metals Science Source Book, p.l. 55. F.R. Beckitt: 'Journals of The Iron and Steel Institute', May 1969, p.632. 56. E.C. Bains and W.E. Griffiths: Transactions AIME, vol. 75, 1 927 , p.166. 57. E.O. Hall and S.H. Algie: 'Metallurgical Reviews', vol. 1 1 , 1 966 , p.61 . 58. L. Smith and K.W.J. Bowen: Journal of The Iron and Steel Institute, March 1984, p.295. 59. A.J. Lena: 'Metal Progress', July 1 954 , p.86. 60. C.E. Spaeder, Jr. and K.G. Brickner: Advances in Technology of Stainless Steels', ASTM-STP369 , 1 965 , p.143. 61. R. Blower and G.J. Cox: Journal of Iron and Steel In stitute' , Aug. 1970, p.769. 62. F.H. Beck et al.: 'Advances in Technology of Stainless Steels', ASTM - STP 369, 1965, p.159. 63. E.J. Dulis and G.V. Smith: 'The Nature, Occurrence, and Effects of Sigma Phase', ASTM STP 1 10 , 1 950, p.3. 64. ASM Metals Handbook, 8th Edition, vol. 1, p.419-422. 65. R.J. Gray et al.: 'Journal of Metals', Nov. 1978, p.18. 66. 'Handbook of Stainless Steels', eds. D. Peckner and I.M. Bernstein, published by McGraw-Hill, p.10-10. 67. ' L. Colombier and J. Hochniaun: 'Stainless and Heat Resisting Steels', published by Edward Arnold Ltd., 1965, p.109. 82 68. W. Holzgruber: Proc. 5th International Symposium on ESR, Pittsburgh, eds. G.K. Bhat and A. Simkovitch, 1974, p.70. 69. R.P. DeVries: Proc 2nd International Symposium on ESR, Pittsburgh, 1969. 70. H. Fredriksson: 'Metallurgical Transactions', vol. 3, Nov. 1942 , p. 2989.. 71. H. Fredriksson and 0. Jarleborg: Journal of Metals, Sept. 1971 , p.32. 72. R.O. Jackson and A. Mitchell: 'Journal of Vacuum Science and Technology1, Vol. 9, no. 6, Nov.-Dec. 1972, p.1301. 73. K.J. Irvine et al .: 'The Metallurgical Evolution of Stainless Steels', ASM Metals Science Source Book, p.379. 74. W.T. Pell-Walpole: 'Journal, Institute of Metals', vol. 69, 1943, p.131. 75. ASM Metals Handbook, 9th Edition, vol. 1, p.427-428. 76. ESR 4340 steel data from Cabot Corporation, Texas, 1977. 77. B.E. Paton et al.: Proc. 2nd International Symposium on ESR;, Pittsburgh, 1969. 78. G.K. Bhat: Proc. 3rd International Symposium on ESR, Pittsburgh, 1971, p.241. 79. G.K. Bhat: Proc. 4th International Symposium on ESR, Japan, June 1973, p.196. 80. G. Hoyle: 'Electroslag Refining', Iron and Steel Institute, London, 1973, p.136. 81. B.E. Paton et al.: 'Special Electrometallurgy', 1972, Kiev, p.174. 82a TABLES TA8i_E_I_ Rene'iting Log for Stainless Steel ESC Valves Valve No. 5 Valve Mo. 6 Valve No. 7 Valve No. 9 . Valve No. 10 Electrode Material CF-8M Rolled 316 Rolled 316 CF-8M Rolled 316 Electrode Diameter 3.5" (88.9 mm) 3.0" (76.2 mm) . 3.0" (76.2 mm) 3.5" (88.9 mm) 3.0" (76.2 mm) Slag Composition 702CaF2/15%Al203/15%Ca0 70SCaFo/15%A!.„0,/15XCa0 70«CaF2/l 5%A)i,03/15XCaO 70«CaF2/15XAi203/"i 5SCa0 702CaF2/15ftU,u3/153!CaO Slag Weight 6.8 kgs 6.8 kgs 6.6 kgs 6.8 kgs 6.8 kgs Slag Condition Hot, dry, prefused Hot, dry, prefused. Hot, dry, prefused . Hot and dry Hot and dry Average Secondary Current 3.72 KA 2.41 KA 3.50 KA 4.04 KA 3.71 KA Average Voltage 37.1V 38.0 V 38.0 V 33.4 V 37.0 V Average Melt Rate C.942 kg/min 0.833 kg/min 0.868 kg/min 0.866 kg/min 0.888 kg/min — Total Malt Tir.e 88 mins 101 mins 33 mins 117 mins 97 mins Hot Topping Cycle None 6.8 mins at 2.3 KA 6.9 mins at 2.1 KA 6.1 mins at 2.2 KA 5 mins at 2.2 KA Deoxidant A? powder Aa powder As. powder A2 powder Al powder Rate of Deoxidation 1.2 grns/min 1.5 gms/min 1.6 gms/min 1.2 gms/min 1.3 gms/min Heltino Atmosphere Air Partial Argon Partial Argon Partial Argon Partial Argon Al Joying Additions J None None Cr chips at 22.3 gms/mir None FeCr-30;4 gms/min FeMo-<4.5 gms/min CO OJ 84 -TABLE II Ferrite Numbers of Stainless Steel Castings Valve No. 5 - CF-8M (i) Specimen Location Ferrite Indicator Magne-Gage (F.N.)* As Cast Heat Treated As Cast Heat Treated Top Centre Edge *F.N. - Ferrite 10<F.N.<15 15<F.N.<20 15<F.N.<20 No. 10<F.N.<15 7.5<F.N.<-10 10<F.N.< 15 9.1 16.51 17.18 1.56 10.6 11.5 Valve No. 6 - 316 (ii ) Specimen Ferri te Indicator Magne-Gage (F.N.) Location As Cast Heat Treated As Cast Heat Treated Top < 0.91 0.55 Centre F.N.<2 F.N.<>2 0.161 0.16 Mid-Radius Everywhere Everywhere 0.94 0.43 Edge 0.40 0.47 No. 7 - 316 + Cr (iii) Specimen Ferrite Indicator Magne-Gage (F.N.) Location As Cast Heat Treated As Cast Heat Treated Top 2<F.N.<5 F.N<2 3.34 1.37 Centre 2<F.N.<5 5<F.N.<7.5 5.56 5.29 Bottom 5<F.N.<7.5 5<F.N.<7.5 4.31 5.75 Mid-Radius 5<F.N.<7.5 5<F.N.<7.5 4.72 5.05 Edge 5<F.N.<7.5 2<F.N.<5 5.68 3.28 85 TABLE II Ferrite Numbers of Stainless Steel Castings (Continued) Valve No CF-8M (iv) Specimen Location Ferrite Indicator Magne-Gage (F.N.) As Cast Heat Treated As Cast Heat Treated Top Centre Bottom Mid-Radius Edge 5<F.N.<7.5 15<F.N.<20 15<F.N.<20 15<F.N.<20 15<F.N.<20 5£F.N<.<7.5 15<F.N.<20 10<F.N.<15 20<F.N.<25 15<F.N.<20 6.33 14.85 14.71 16.09 14.75 5.00 15.35 12.62 19.46 15.81 Conventional Casting - CF-8M (v) Specimen Location Ferrite Indicator Magne-Gage (F.N.) Heat Ireated Heat Ireated Top 3/4 Height Centre 1/4 Height Bottom Mid-Radius Edge 20'<F.N.<25 20<F.N.<25 20<F.N.<25 20<F.N.<25 F<N.<25 15<F.N.<20 20<F.N.<25 20.57 20.67 20.45 18.99 26.78 16.99 28.70 TABLE III Average Ferrite Numbers of Stainless Steel Castings Determined by Magne-Gage. and Schoefer's Diagram Casting As Cast Heat Treated* Schoefer's Diagram Valve No.5 - (CF-8M) 16.85 10.80 16.8 Valve No.6 - (316) 0.65 0.35 <-!0 Valve No.7 - (316+Cr) 5.07 4.84 4.5 Valve No.9 - (CF-8M) 15.10 15.81 22 Conv.Cast - (CF-8M) - 19.53 32 *From Magne-Gage 86 TABLE IV Interdentriti c Mi crosegregation Ratios of CF-8M Castings Casting and Pos i ti on Segregation Ratio (C max/C min) Chromi urn Ni ckel Molybdenum Valve No. 5* (CF-8M) Centre Valve No. 9* (CF-8M) Edge Centre Conv. Casting**(CF-8M) Edge Mi d-Radi us Centre 1 .349 1 .25 1.18 1 .167 1 .305 2.337 2. 37 2.19 2.48 1 .625 2.178 2.587 2.10 1 .74 2.60 3. 568 3.859 3.299 * in as-cast condition ** in solution-treated condition 87 TABLE V Chemical.Composition of Valve No. 5 (CF-8M) (wt.SS) C Mn Si P S Ni Cr Cu Mo hi Nb Co Ti Zr Electrode 0.066 0.33 1.19 0.027 0.016 10.19 20.31 0.44 2.51 - 0.073 0.108 0.008 0.006 Top 0.086 0.34 1.06 0.021 0.010 10.08 19.48 0.37 2.55 - 0.0C5 0.115 NA* NA* Centre 0.063 0.26 0.74 0.019 0.010 9.80 19.86 0.15 2.53 - 0.060 0.150 11 II Side (L) 0.055 0.25 0.73 0.019 0.009 9.88 19.39 0.16 2.55 - 0.060 0.150 II II Side (R) 0.056 0.26 0.73 0.220 0.010 9.91 19.72 0.16 2.58 • - 0.061 0.158 ••• II N.A. - not analysed TABLE VI Chemical Composition of Value No. 6 (316) '(wt'.*) C Mn Si , P s Ni .Cr . . Cu . Mo- A£ . Nb Co Ti Zr Electrode 0.049 1.89 0.67 0.030 0.031 11.60 17.56 0.09 1.98 - 0.062 0.445 0.008 0.006 Top 0.072 1.89 0.64 0.031 0.011 11.91 16.84 0.13 2.11 - 0.061 0.278 N.A.* N.A.* Centre 0.073 1.85 0.64 0.030 0.011 11.80 16.89 0.12 2.09 - 0.059 0.266 II » Bottom 0.075 .1.84 0.64 0.033 0.011 11.65 17.09 0.12 2.12 - 0.059' 0.263 " II Side (L) 0.069 1.84 0.63 0.034 0.011 11.42 16.74 0.12 2.05 - 0.061 0.258 II 11 Side.(R) 0.070 . 1.86 0.65 0.031 0.011 11.80 17.08 0.13 2.08 - 0.060 0.283 II II *N.A. - not analysed 88 TAG l E VII Chemical Composition of Valve No, 7 (316-tCr ) (w t. %) C tin Si P , S K i | C r Cu Ko . A i Nb | Co Ti Zr Electrode 0.019 1.S9 0.67 0.030 0.031 11.60 17.. 56 0.09 1 .S3 - 0.052 0.195 0.008 0.006 Top 0.074 1.87 0.61 0.035 0.012 1 .18 18.91 0.12 2.01 - 0.061 0.285 (.'.A.* N.A.* Centre 0.078 1.80 0.58 0.029-0.012 n.37' IS.93 O.o2 1.98 - 0.61 0.261 Bottom 0.072 1.79 0.5-5 0.036 0.011 11.13 19.25 0.12 2.00 0.60 0.256 II » Side (R) 0.062 1.81 0.57 0.056 0.013 11.26 19.88 0.12 2.01 - 0.61 0.290 -*N.A. - not analysed TABLE VIII Chemical Composition of Valve Ho. 9 (CF-8H) (wt.,%) C Hn Si P S Ni Cr Cu Mo At Nb Co Ti Zr Electrode 0.066 0.33 1.19 0.027 0.016 19.19 20.31 0.11 2.51 - 0.073 0.108 0.008 0.006-Top 0.058 0.35 1.01 0.028 0.009 10.33 19.87 0.10 2.51 - 0.068 0.089 0.013 0.005 "Centre 0.036 0.19 0.66 0.018 0.009 10.15 19.67 0.13 2.55 - 0.060 0.159 0.005 0.001 Bottom 0.037 0.27 0.67 0.021 0.009 9.73 20.12 0.15 2.11 - 0.063 0.176 0.005 0.006 Side (L) 0.039 0.21 0.72 0.021 0.009 9.63 20.51 0.13 2.19 - 0.061 0.185 0.005 0.006 Side (R) 0.011 0.20 0.70 0.020 0.009 9.90 20.19 0.13 2.51 - C.063 0.193 0.001 0.006 89.. TABLE IX Chemical Composition of Valve No. 10 (316+Cr+Mo) {vit.%) C Mn Si P S Ni Cr Cu Mo At Nb Co Ti Zr Electrode 0.049 1.89 0.67 0.030 0.031-11.60 17.56 0.09 1.98 - 0.062 0.445 0.008 0.004 Top 0.055 1.69 0.43 0.030 0.010 12.10 18.47 0.10 2.32 - 0.063 0.102 0.007 0.005 Centre 0.046 1.65 0.41 0.027 0.011 11.91 19.14 0.07 2.32 - 0.064 0.134 0.007 0.005 Bottom 0.051 1.62 0.28 0.027 0.009 12.11 18.78 0.10 2.26 - 0.064 0.127 0.007 0.005 Side (L) 0.048 1.65 0.40 0.028 0.011 11.96 19.22 0.08 2.33 - 0.065 0.133 0.007 0.005 Side (R) 0.040 1.61 0.39 0.024 0.010 11.90 19.05 0.09 2.51 - 0.065 0.135 0.007 0.006 TABLE X Chemical Composition of the Conventional Casting (CF-8M) (wt.SS) C Mn Si . P S Ni Cr Cu Mo An Nb Co Ti Zr Top- 0.029 0.61 1.24 0.027 0.010 9.13 19.41 0.25 2.02 - 0.087 0.067 0.009 0.003 Centre 0.023 0.62 1.21 0.026 0.010 8.90 19.80 0.22 2.00 - 0.086 0.071 0.009 0.003 Bottom 0.019 0.65 1.24 0.031 0.010 9.34 19.67 0.23 2.06 - 0.089 0.068 0.008 0.003 Mid-radius 0.024 0.63 1.25 0.029 0.010 9.05 19.85 0.23 2.04 - 0.087 0.072 0.008 0.003 Side (R) 0.020 0.63 1.24 0.031 0.010 9.57 19.36 0.23 2.10 - 0.087 0.066 0.008 0.003 TABLE XI Percent composition of Cr and Mo in areas 1 and 2 and the Parent Valve No. 10 Area Analysed Chromi urn (wt.%) Molybdenum (wt.%) Area 1 54.88 12.56 Area 2 54.21 13.12. Parent 22.10 1 .79 TABLE XII Tensile Properties of Valve Nos. 5, 6 and 7 Casting No. and Elonaation In Yield Strength, Ultimate Tensile Test Direction 1 inch (25.4 mm) 0 2% offset, Strength % Ksi(KPa) Ksi(MPa) Valve No. 5 (CF-8M) Lonaitudinal - 1 45 3 36 (251.7) 77. 2 (532 ..3) - 2 51 3 47 3+ 35 8 (246.8) 37 0f 74 1 (510.9) 77.8+ - 3 45 2 38 6 (266.1 ) 255 \ 82 I (566.4) 536.4 Transverse 1 46 3 36 2 (249.6) 71 7 (494.4) - 2 5 3'. 4 50 39 0 (268.9) 36 91 75 3 (519.2) 72.6 + - 3 50 6 35 5 (244,8) 254 4 70 7 (487.5) 500.6 Valve No. 6 (316) Longitudinal - 1 62 1 34 1 (235.1 ) 71 1 (494.4) - 2 68 8 64 8+ 31 1 (218.6) 32 J. 5' 69 7 (480.6) 68. 3+ - 3 63 6 31 1 (214.4) 224 1 64 1 (442.0) 470.9 Transverse 1 69 6 32 3 (222.7) 68 0 (468.9) - 2 62 4 65 4 + 34 0 (234.4) 32 5+ 69 4 (475.8) 69. 3+ - 3 64 1 31 2 (215.1) 224 1 70 6 (486.8) 477.8 Valve No. 7 (316+Cr) Longitudinal - 1 51 1 34 9 (240.6) 73 6 (507.5) 71.7 + _ 2 68.6 57 J. 9 ' 35. 2 (242.7) 34 7 + 68 .4 (471.6) 494.4 • - 3 54 0 33 9 (233.7) 239 3 73 2 (504.7) Transverse 1 56 0 ' 34 0 (234.4) 66 3 (457.1) - 2 53 4 61 3+ 39 0 (268.9) 35 7 + 75 3 (519.2) 70.8+ - 3 74 5 34 2 (235.8) 246 2 70 7 (487.5) 488.2 + Average value. TABLE XIII Tensile Properties of Valve No. 9, Conventional Casting and ASME/ASTM Standards for AISI 316 and ACI CF-8H Casting No. and Test Direction Reduction In Area, % Elongation In 1 Inch (25.4 mm) X Yield Strength, 0.2% offset, Ks1 (MPa) Ultimate Tensile Strength Ks1 (MPa) .Valve No. 9 (CF-8M) Lonqltudlnal - 1 80.0 70.7 39.3 (271.0) 82.4 (568.1) - 2 82.0 Cl.O* 71.3 71.0f 42.4 (292.3) 40.9+ 82.1 (566.1) 82.3f 567.5 Transverse - 1 82.4 70.0 41.8 (288.2) 282.0 82.3 (567.5) - 2 - 3 78.7 77.9 73.0 71.2* 70.5 42.1 (290.3) 42.6 (293.7) 42.4+ 291.0 81.1 (559.2) 83.2 (573.7) 82.2+ 566.8 Edge - 1 79.4 73.7 44.2 (304.8) 86.3 (595.0) - 2 82.0 80.7* . 69.0 71.4f 42.8 (Z95.1) 43.5+ 315070" 84.5 (582.6) 85.4+ 588.8 t Average value. Conventional Cast (CF-8M) Lonqltudlnal* -0.25 Inches from the edge 74.3 46.6 57.0 (393.0) 82.5 (568.8) 2.50 " » - • 72.6 52.0 45.0 (310.3) 82.0 (565.4) 4.25 " ' " " " 70.1 58.9 46.7 (322.0) 85.2 (587.5) 6.00 " " " " 62.0 53.Z 46.7 (322.0) 85.8 (591.6) 7.75 " " " " 63.4 56.7 45.9 (316.5) 84.6 (583.3) Transverse* 1.90 Inches from the edge 72.2 58.3 46.4 (319.9) 84.5 (582.6) 3.75 « . . 75.2 55.5 46.0 (317.2) 83.2 (573.7) 5.60 " " " " 65.7 55.3 46.6 (321.3) 84.7 (584.0) * Each value represents an average of 3 tests. ASME SA-182 or ASTM A-18Z-77a(316) 50 (m1n) 30 (min)* 30 (207) m1n 70 (483) min ASME SA-351 or ASTM A-351r77(CF-8M) not specified 30 (min)* . 30 (207) m1n 70 (483) m1n * Elongation 1n 2 Inches (50.8 nm) 10 TABLE XIV. Tensile Properties of Large Specimens from Valve Nos. 6 and 9 and Small Specimens from a Large Specimen. Casting No. and Test Direction Reduction In Area, % Elongation In 1 inch (25.4 mm) % Yield Strength 0.2% offset, Ksi (MPa) Ultimate Tensile Strength Ksi (MPa) Valve No. 6 (316) Longitudinal - 1 - 2 - 3 Valve No. 9 (CF-8M) Longitudinal - 1 - 2 78.8 82.0 81.3f 83.0 85.2 80.7 83.0+ 70.9 69.3 66.0+ 57.8 47.9 47.5 47.7+ 32.6 (224.8) 29.0 (200.0) 30.9+ 31.0 (213.7) 213-1 \ 36.2 (249.6) 36.7+ 37.1 (255.8) 253-0 66.8 (460.6) 61.8 (426.1) 63.7+ 62.6 (431.6) 439-2 73.0 (503.3) 73.5+ 74.0 (510.2) 506-8 Valve No. 9 (CF-8M) (From large Longitudinal -1 Edge - 1 Edge - 2 Centre - 1 87.0 87.1 86 24.2 • 35.8 73.1 71.1 + 84.5 38.7 (266.8) 38.4 (264.8) 38.0+ 37.0 (255.1) 262-° 78.2 (539.2) 73.0 (503.3) 75.0+ 73.7 (508.2) 517-1 t Average value. TABLE XV Remelting Log for AISI 4340 Steel ESC Valves Valve No. 3 Valve No. 8 Valve No. 11 Valve No. 12 Valve No. 13. Valve No. 14 Electrode Materia"! Rolled 4340 Rolled 4340 Rolled 4340 Soiled 4340 Rolled 4340 Rolled 4340 Electrode Diameter 3.25" (82.6 mm) 3.25" (82.6 mm) 3.25" (82.6 mm) 3.25" (82.6 mm) 3.25" (82.6 mm) 3.25" (82.6 mm) Slag Composition 61.75%CaF2/26.5%M203/ 11.75%Ca0 46%CaF2/17»203/ 17%Ca0/20%La203 46%CaF2/17«At203/ 17%Ca0/20%La203 46XCaF2/mAt203/ 17SSCa0/202La203 46%CaF2/172At203/ mCaO/20XLa203 462CaF2/173!Ai203/ 173XaO/20XLa203 Slag Weight 9.1 kg 6.8 kg 6.8 kg 6.8 kg 6.8 kg 6.8 kg Slag Condition Liquid CaF„, Hot A«203 and CaO Hot and Dry Hot and Dry Hot and Dry . Hot and Dry Hot and Dry Average Secondary | Current 2.73 KA 3.72 KA 3.59 KA 3.60 KA 3.55 KA 3.60 KA j Average ' 1 Voltage 36.9 V 37.1 V 36.9 V 36.8 V 35.4 V 36.0 V Average Melt Rate 1.022 kg/min 0.938 kg/mins . 0.967 kg/min 0.854 kg/min 0.785 kg/min 0.812 kg/min Total Melt Time 87 mins 98 mins 98 mins • 112 mins 113 mins 110 mins Hot Topping Cycle 3.2 mins at 2.6 KA 7.2 mins at 2.2 KA 6.7- mins at 2.2 KA 8.3 mins at 2.2 KA -6.7 mins at 2.2 KA 9. mins at 2.3 KA . Deoxidant At powder At powder At powder At powder At powder At powder Rate of Oeoxidation 2.4 gms/min 1.2 gms/min 1.2 gms/min 1.2 gms/min 1.4 gms/min 1.4 gms/min Melting |Atmosphere Air Partial Argon Partial Argon' Partial Argon Air Air OJ 94 TABLE XVI Interdendritic Microsegregation at the Centre of AISI 4340 ESC Valves Val ve No. Segregation Ratfo (C m ax/C min) Chromium Nickel Molybdenum Valve No. 3* Valve No. 8* Valve No. 13** (At mid-radius) Valve No. 14** 1 .39 1 .34 1.12 1 .30 1.13 1.14 1 .60 1 .05 2.15 2.21 1 .24 *In as-cast condition **The whole valve was heat-treated. TABLE XVII Chemical Composition of Valve No. 3 (4340) (wt.%) C Mn Si P S Ni Cr Cu Mo An Nb Co V B W Ti Zr Electrode 0.42 0.77 0.37 0.021 0.015 1.87 0.80 0.10 0.21 0.013 0.053 0.029 0.012 0.0006 0.018 0.011 0.003 Top 0.42 0.78 0.41 0.021 0.004 1.84 0.83 0.07 0.22 0.095 0.024 0.033 0.017 0.0014 0.026 N.A. N.A. Centre 0.40 0.76 ' 0.40 0.019 0.004 1.80 0.81 0.08 0.21 0.082 0.023 0.031 0.016 0.0011 0.026 II " Bottom 0.40 0.73 0.37 0.018 0.004 1.79 0.80 0.08 0.21 0.127 0.020 0.028 0.016 0.0013 0.024, II Side (L) 0.40 0.75 0.38 0.018 0.003 1.81 0.81 0.08 0.21 0.085 0.0H 0.028 0.016 0.0009 0.022 « •• Side (R) 0.42 0.78 0.40 0.019 0.003 1.83 0.82 0.08 0.22 0.073 0.02; 0.031 0.016' i 0.0010 0.024 *N.A. - not analysed CO TABLE XV11I Chemical Composition of Valve No. 8 (4340) (wt.%) C Mn Si P S Ni Cr Cu Mo M Nb Co V B w Ti Zr Electrode 0.42 0.77 0.37 0.021 0.015 1.87 0.80 0.10 0.21 0.031 0.053 0.029 0.012 0.0006 0.018 0.011 0.003 Top 0.43 0.80 0.33 0.030 0.004 1.90 0.81 0.11 0.22 0.035 0.056 0.042 0.013 0.0007 0.024 0.005 0.004 Centre 0.41 0.77 0.34 0.028 0.004 1.86 0.80 0.11 0.22 0.048 0.056 0.039 0.012 0.0007 0.023 0.005 0.004 Bottom 0.40 0.75 0.20 0.29 0.003 1.74 0.76 0..10 0.21 0.051 0.052 0.030 0.012 0.0005 0.016 0.004 0.003 Side (L) 0.41 0.75 0.34 0.29 0.003 1.80 0.78 0.10 0.21 0.043 0.053 0.032 0.012 0.0007 0.016 0.005 0.004 Side (R) 0 '.-41 0.79 0.34 0.28 0.002 1.87 0.81 0.11 0.22 0.045 0.051 0.024 0.012 0.0006 0.019 0.004 0.003 LO CXl TABLE XIX Chemical Composition of Valve No. 13 (1310) (wt.%) C Mn Si P S Ni Cr Cu Mo At Nb Co V • B W Ti Zr Electrode 0.42 0.77 0.37 0.021 0.015 •1.87 0.80 0.10 0.21 0.031 0.053 0.029 0.012 0.006 0.018 0.011 0.003 Top 0.43 0.84 0.18 0.030 0.003 1.89 0.86 0.31 0.21 0.020 0.053 0.055 0.011 0.0004 0.041 0.003 0.002 Centre 0.42 0.86 0.11 0.030 0.001 1.88 0.86 0.3o 0.21 0.021 0.053 0.057 0.011 0.0004 0.044 0.002 0.002 Bottom 0.42 0.76 0.04 0.030 0.001 1.79 0.83 0.30 0.21 0.017 0.052 0.059 0.010 0.0004 0.039 0.003 0.003 Side (L) 0.40 0.80 0.09 0.029 0.002 1.75 0.82 0.31 0.20 0.021 0.052 0.062 0.011 0.0004 0.033 0.004 0.003 Side (R) 0.41 0.82 0.10 0.030 0.002 1.80 0.84 0.31 0.21 0.022 0.055 0.065 0.011 0.0004 0.037 0.004 0.003 TABLE XX Chemical Composition of Valve No. 1 4 (4340) (wf.%) C Mn Si P S Ni Cr Cu Mo At Nb Co, V B w Ti Zr Electrode 0.42 0.77 0.37 0.021 0.015 1.87 0.84 0.10 0.21 0.031 0.053 0.029 0.012 0.0006 0.018 0.011 0.003 Top 0.44 0.84 0.19 0.033 0.004 1.90 0.86 0.31 0.21 0.018 0.058 0.071 0.012 0.0006 0.050 0.003 0.004 Centre 0.45 0.88 0.14 0.032 0.002 1.98 0.88 0.30 0.22 0.022 0.057 0.066 0.012 0.0006 0.056 0.001 0.003 Bottom 0.46 0.87 0.05 0.033 0.002 2.03 9.90 0.30 0.22 0.013 0.059 0.072 0.012 0.0005 0.064 0.001 0.004 Side (L) 0.45 0.89 0.09 0.033 0.002 1.94 0.87 0.31 0.22 0.018 0.059 0.088 0.013 0.0006 0.050 0.003 0.004 Side (R) 0.45 0.89 0.09 0.032 0.002 1.97 0.88 0.30 0.22 0.016 0.060 0.090 0.012 0.0006 0.055 0.001 0.004 to 98 TABLE XXI Sulphur Contents of AISI 4340 Electrode and ESC Valves Electrode Valve No. 3 Valve No. 8 Valve No. 13 Valve No. 14 Sulphur Content (wt.%) 0.0146 0.0040 0.0037 0.0055 0.0053 TABLE XXII Dimensional Measurements on Valve Nos. 11, 12 and 13 Val ve No. Shrinkage Type % Shrinkage and Location ; a b c d e f 9 h i Valve No.11 Real Pseudo 2.37 3.60 2.68 4.10 2.33 3.75 1.35 3.05 1.97 3.09 2.55 4.55 1.48 4.34 1.67 4.17 1.92 4.12 Valve No.12 Real Pseudo 1.87 3.36 1.07 2.31 1.72 2.80 1 .29 2.70 2.34 3.26 0.47 1.73 0.97 2.22 1.27 2.52 1.61 2.87 Valve No. 13 Real Pseudo 2.43 4.64 2.01 4.26 2.53 4.25 2.36 4.22 2.27 3.30 . 2.18 3.36 2.23 3.30 2.19 3.37 2.99 4.17 TABLE XXIII Tensile Properties of AISI 4340 Electroslag Cast Valves Casting No. and Test Direction Valve No. 3 (a) Longitudinal -Transverse Valve No. 8 (b) Longitudinal -Transverse Edge Valve No. 13 (c) Longitudinal -Transverse Valve No. 13 (d) Longitudinal -Transverse Reduction 1n Area, % 48.0 33.3 30.1 33.9 37.8 36.4 46.0 42.2 47.7 41.8 44.2 44.7 51.3 46.1 7.1 9.4 10.9 11.7 8.7 9.4 39.9 42.0 32.9 31.7 37.11 36.0T 45.3T 43.61 48.7 9.1' 9.9 + 41.0T 32.3T Elongation In 1 inch (25.4 mm), 10.1 7.2 8.5 9.3 11.2 8.4 16.5 14.7 17.9 15.5 17.0 17.8 14.1 17.5 3.7 4.2 5.8 3.5 4.9 4.5 16.7 18.5 17.0 13.3 8.6+ 9.1 + 16.4+ 16.8T 15.8T 4.6 + 4.3 + 17.6T 15.2T Yield Strength 0.2% offset, Ksi (MPa) 174.6 (1197.0) 176.8 (1212.1) 176.2+ 179.1 (1234.9) 1214'9 174.0 (119917) 171.8 (1184.6) 172.7+ 172.9 (1187.3) 1190,8 149.9 (1033.6) 149.3 (1029.4) 140.01 150.7 (1039.1) 1034'3 151.8 (1046.7) 153.0 (1054.9) 152.6 (1052.2) 105K5 152. $ 149.1 (1028.0) 150.5 (1037.7) 149.8 1032.9 120.1 120.1 (828.1) (828.1) 120.1 (828.1) 122.4 (843.9) 120.1 (828.1) 120.1 (828.1) 147.7 (1018.4) 147.3 (1015.6) 146.2 (1008.0) 144.3 (994.9) 120. V 828.1 120.tf 833.6 147.5+ 1001.8 Ultimate Tensile Strength Ksi (MPa) 185.1 (1276.3 187.8 (1292.8 188.0 (1296.3 184.1 (1269.4 184.7 (1273.5 182.7 (1259.7 164.7 (1130.8 164.3 (1132.8 165.1 (1138.4 166.3 (1146.5 167.2 (1152.8 165.8 (1143.2 160.4 (1106.0 165.2 (1139.1 142.1 138.8 136.6 134.1 135.4 136.8 (979.8 (957.0 (941.9 (924.6 (933.6 (943.2 159.8 (1101.8 1017.0| 15g3 (1098.4 145.3+1 157.9 (1088.7 155.3 (1070.8 187.0 + 1289.4 183.8+ 1267.3 164.7+ 1135.6 166.4' 1147.3 162.8 1122.5 139.2 959.8 135.4+ 933.6 159.6T 1100.4 156.6+ 1079.8 (a) Tempered at 482°C to a hardness of 38 Rc. (bj Tempered at 550°C to a hardness of 35 Rc. (c) The whole valve was heat-treated and tempered at 560°C. Average hardness of the tensile specimens was 31 Rc. (d) The specimen bars were re-heat-treated and tempered at 560°C to a hardness of 33 Rc. t Average value. CD CD TABLE XXIV Tensile Properties of Conventional AISI 4340 Hot Rolled Bar Specimen No. and Test Direction Reduction In Area, Elongation In 1 inch (25.4 mm) Yield Strength at Lower Yield Point Ksi (MPa) Ultimate Tensile Strength Ksi (MPa) Lonqitudinal - 1 - 2 - 3 Transverse - 1 - 2 - 3 57.7 58.2 57.8+ 57.5 21.9 22,7 22.7f 23.5 19.2 19.5 18.3f 16.2 9.4 9.5 9JS+ 10.5 150.7 (1039.1) 150.0 (1034.3) 150.8+ 151.8 (1046.7) 1039'8 150.3 (1036.3) 149.1 (1028.0) 150.2+ 151.3 (1043.2) 1035-6 162.8 (1122.5) 164.3 (1132.8) 163.61" 163.6 (1128.0) 1,28'° 162.0 (1117.0) 161.1 (1110.8) 162.0+ 162.8 (1122.5) lll7-° (a) Tempered at 560°C to a hardness of 33.5 Rc' Diameter of the bar was 3.25 inches (82.6 mm) t Average value. TABLE XXV Transverse Tensile Properties of Air Melted and Vacuum Arc Melted 4340 Steels Tempering Temperature Reduction % in Area, Elongation In , 2 inches (50.8 mm), % Yield Strength Ksi (MPa) Tensile Strength Ksi (MPa) Air Melted 480°C 14 9 8.0 173 (1192.8) 200 (1379) 540°C 22 0 10.0 163 (1123.9) 180 (1241.1) Vacuum Arc Remelted 480°C 20 0 9.0 175 (1206.6) 200 (1379) 540°C 24 0 10.5 160 (1103.2) 180 (1241.1) * Properties listed are averages of several heats from the same producer; billet size and amount of hot redi were not available. TABLE XXVI Lonqitudinal Mechanical Properties of Bar Stock Made From Remelted 4340 Steel Melting Method Reduction In Area, % Elongation In 4D. % Yield Strength Ksi (MPa) Tensile Strength Ksi (MPa) Hardness (MRc) VAR<b> ESR<C> 61.2 59.0 16.4 16.1 163 (1123.9) 158 (1089.4) 175 (1206.6) 171 (1179.0) 37 37 (a) Bars were normalised at 900°C, oil quenched from 845°C, and tempered 2 hrs. at 541°C. All specimens taken from mid-radius. (b) 3.62 inch round. (c) 4.625 inch round. TABLE XXVII Mechanical Properties of ESR 4340 Material in the Transverse Direction Heat-Treated to Different Strength Levels.*76 Tempering Temperature Reduction in Area, % Elongation - % 0.2* Yield Strength Ksi (MPa) Tensile Strength Ksi (MPa) 538°C 566°C 45.1 49.2 12.9 14.6 169.4 (1168.0) 158.3 (1091.5) 180.6 (1245.2) 173.0 (1192.8) * Data from Cabot Corporation. Notes (1) Tensile data were generated from a 0.357 in. (9mm) diameter test fears machined from heat-treated 1-in. diameter transverse bars taken from a. 24-in. diameter ESR ingot forged 3:1 to 14-in. diameter. (2) All samples were normalised at 1750°F (954°C) prior to heat-treating, then austenitized at 1525°F (830°C) and tempered at various temperatures. (3) Tensile values are average of five tests. TABLE XXVII 1 (a) Tensile Properties of Large Specimen from ESC Valve of AISI 4340 Steel. Casting No. and Test Direction Reduction In Area, % Elongation In 3 inches (76.2 mm) % Yield Strength, 0.2% offset, Ksi (MPa) Ultimate Tensile Strength, Ksi (MPa) Valve No. 8* Longitudinal - 1 - 2 48.1. 47.9 1 47.6 19.7 18.11' 16.5 144.9 (999.1 ) 145.5'1' 146.0 (1006.7) 1003-2 160.6 (1107.3) 160.5+ 160.3 (1105.3) 1106-6 * Tempered at 550°C to a hardness of 34.5 Rc. + Average value. TARIF XXVITTIM Tensile Properties of Small Specimens Cut From Large Tensile Specimens Specimen No. and Test Direction Reduction In Area, % Elongation In 1 inch (25.4 mm) % Yield Strength, 0.2% offset, Ksi (MPa) Ultimate Tensile Strength, Ksi (MPa) Value No. 8 (From Large Longitudinal -1) Edge - 1 Edge - 2 Centre r 1 45.2 48.6 46.5f 45.7 16.7 17.4 17.41 18.0 149.7 (1032.2) 144.7 (997.7) 146.31' 144.5 (996.3) 1008'7 164.8 (1136.3) 159.7 (1101.1) 161.4 + 111? 9 159.7 (1101.1) "U o + Average value. 102 TABLE XXIX FATT Values Estimated from Ductile Brittle Transition Characteristics of AISI- 4340 ESC Valves and Electrode, Type and Orientation of Specimen Est i mated FATT Valve No. 3 -- Long. -40°C Electrode - Long. - Trans. 5°C 45°C Valve No. 8 • - Long. . - Trans. - Trans. - Edge (T) (L) -25°C -30°C -35°C -32°C Valve No. 1 3 -- Long. - Trans. 20°C 20°C Notes: (1) The FATT values have been estimated according to a suggested method in ASTM E23-72. (2) FATT has been determined as the temperature corresponding to the energy value 50% of the difference between values obtained at 100% and 0% fibrous fracture. (3) The specimens have been assumed to be 100% fibrous at 100°C and 0% fibrous at -100°C. The latter is not exactly true as some areas were found to be ductile even at -100°C. (4) The high FATT for Valve No. 13 is probably due to the inadequate heat treatment. (5) These are just estimated values and not exact because of the kind of fractured surface (as mentioned in the text). 100a FIGURES \ (d) Figure 1. Electroslag Cast Products.8 (a) Valve Bodies; (b) Rolls; (c) Dentures (d) Crankshaft with pin diameter of 480 mm. 104 1 straight tube, 2 source material, 3 ring-type mould, 4 molten-slag bath, 5 molten-metal pool, 6 slag film, 7 start piece, 8 feed roller for source material, 9 drawing apparatus, 10 electric power source, 11 cool ing water Figure 2. 'Y0Z0' process of Mitsubishi Heavy Industries Ltd: 11 12 products pi pe , tube. ' ELECTRODE ^ . „ „ „. i , POWER SUPPLY TRANSFORMER or RECTIFIER X SLAG POOL METAL POOL SOLIDIFIED CASTING SLAG SKIN COOLING WATER CRUCIBLE BASE PLATE Figure 3. Schematic of the ESC process o Figure 4. Fracture toughness of ESC AISI 4340 steel 2 containing varying amounts of sulphur. Figure 5. Electroslag casting installation at U.B.C. 108 Figure 6. Orthogonal views of ESC valve body. 109 Figure 7(a). Plan views of the valve mold segments. 110 Section A-A Section A'-A1 Figure 7(b). Sections through the valve mold segments. Figure 9. Slag melting furnace. Figure 10. Solidified slag skin on the casting. 113 Figure 12. Stainless steel ESC valve. 115 Figure 13. Longitudinal holes in the castings due to moisture. Figure 14. Schematic of the sectioning procedure of the valves. Figure 15. Macrostructure of Valve No. 5 (CF-8M) Figure 16. Macrostructure of Valve No. 6 (316). 118 Figure 17. Macrostructure of Valve No. 9 (CF-8M). Figure 18. Macrostructure of Valve No. 10 (316+Cr+Mo). 120 Figure 19. Macrostructure of conventional casting (CF-8M). 640 x Figure 20(a). Microstructure of CF-8M ESC valves before heat treatment (etchant - oxalic acid). 122 640 x Figure 20(b). Microstructure of CF-8M ESC valves after heat treatment (etchant - oxalic acid). 210 x 540 x 1100 x Figure 21. Microstructure of as-cast ESC Valve Nos. 5, 6 and 7 (etchant - oxalic acid). Top Row - Valve No. 5 (CF-8M), Middle Row - Valve No. 6 (316), Bottom Row - Valve No. 7 (316+Cr). 124 210 x 540 x 1100 x Figure 22. Microstructure of heat-treated ESC Valve Nos. 5, 6 and 7 (echant - oxalic acid). Top Row - Valve No. 5 (CF-8M), Middle Row - Valve No. 6 (316) Bottom Row - Valve No. 7 (316+Cr). 125 640 x 1300 x Figure 23. Microstructure of CF-8M specimen heated to 850°C (etchant - KOH). light grey area - austenite phase dark grey area - ferrite phase black area - sigma phase Figure 24. Microstructure of heat-treated CF-8M conventional casting (etchant - KOH). 127 Figure 25. Microstructure of Valve No. 9 in as-cast condition (etchant - KOH). 128 Figure 26. Microstructure of Valve No. 9 in heat-treated condition (etchant - KOH). 700 x Figure 27. Microhardness indentations on sigma and austenite phases. Figure 28. Schoefer's diagram for determination of ferrite numbers of stainless steel castings. Figure 29(a). Variation of Cr, Ni and Mo across the dendritic direction in Valve No. 9 at the edge. 132 • TRAVEL t H 1-0 mm Figure 29(b) Variation of Cr, Ni and Mo across the dendritic direction in Valve No. 9 at the centre. 135 • TRAVEL i 1 10 mm. direction in conventional casting at the centre O.II6| 0.066 0.016 2.65 2.55 2.45 10.41 9.91 9.41 20.04 19.54 19.04 • • • • • • o 0 0 o 0 —© —8— O —A —0— w —9 • o • • o o o-6 O o 0 8 9 10 II .7 12 13 14 15 8 195 mm. 10 I 2 3 4 / Microprobe 12 13 14 15 C Mo Ni Cr E e in _i_ _i_ _i_ ^ ^1" — iDiOuD O IT) O ^. °! ^. ifilD — lO — oicrib o^°^Q cJcvJcvj ri ° °. CM O Figure 31. Composition variation in Valve No. 5 (CF-8M). (The solid line shows the overall average composition.) • • 9 • • • -• • o • • • • • o o m o • o o o O o o o —o o u • • \j • • • • • • • # W W 0 o o o o o o o o -0 0 o o U -10 II 12 13 14 15 16 3 17 18 19 20 21 22 23 10 0.089 0.069 0.049 2.13 2.08 2.03 12.19 11.69 11.19 1732 16.82 16.32 295 mm-12 13 14 15 16 17 18 19 20 21 22 23 cn <r> cr> ^ & 00 CD 01 m rOCOro srtoco roCOrO ^ <X> 22 Q C2 OQQ ^> ^ t: r = ™ c\i oj <\i odd o 0 Figure 32. Composition variation in Valve No. 6 (316) (The solid line shows the overall average composition.) 138 0.094 0.069 0.044 2.05 ZOO 1.95 11.86 11.36 10.86 19.62 19.12 18.62 - o o o o o o o 6 • • * • • _ • o u CJ o o - • • • • • • • • 4 II 12 13 14 15 16 17 c Mo Ni Cr 8 9 10 -155 mm. 12 13 14 16 17 e £ m O CVJ 8 9 10 Cr Ni Mo C C\JC\JC\I co ID co LQC; in "tf cn <3-CO _ CD COrOOO CT) CJ O <X> <X> crjcricT) o—— — ^ ^ ooo boo Figure 33. Composition variation in Valve No. 7 (316+Cr). (The solid line shows the overall average composition.) 139 0-061 0041 0021 263 2-51 239 10-49 9-99 949 20-48 19 98 19-48 --o o o 0 o o ^ n n 0 v—O— 6 o o • • • ° n o - o 0 o o I9|2C|2I |22|23|24|25ll0|26 27 28 29 30|3I 32 Mo Ni Cr microprobe -295 mm. Cr Ni Mo j2\|22|23l24|25 10 26|27|28|29l30|3l|32 E E in N ro 1 o • 2 • 3 > • 0 • 4 0 • o • 5 o m o • 6 < o • 7 0 0 • e 0 o • 9 o o • 10 ( 0 • 11 0 0 • 12 o a • 13 0 > • 14 o o 15 3 • o 16 0 • 0 17 0 • 0 18 I- . ? t i 1 oo oo oo cn cn cn ^- cn ? t cn if ch w o ch cn o £ ft X o do o oo Figure 34. Composition variation in Valve No. 9 (CF-8M). (The solid line shows the overall average composition.) 140 0-059 0O49 0-039 289 239 I- 89 12-56 12-06 II- 56 1940 18-90 18-40 - • o • 0 -A • • • W w • - 0 O o o u 0 o O u - n • • • • • • • • • - o Q 0 o "D o o o -o o u 15 1-6 17 18 19 6 20 21 22 23 24 IQR mrr Mo NI Cr 15 16 17 18 19 10 I I 12 13 14 20 21 22 23 24 E 6 ID Cr NI Mo 1 o • < > • 2 o c » 3 o o • 4 c » c ( » 5 < • 0 • 6 o • 1 ) • 7 o « c • 8 o « c ( 9 o • c • 10 o • c < 1 1 0 • 0 < 12 c • o • 13 ) • 0 • 14 0 1 1 1 * 1 1< 1 1 1( »1 OOO <£<0<£ ©0105 to*- "Opip cbroeo cb cb — — OJ men g> oo o 66 6 Figure 35. Composition variation in Valve No. 10 (316+Cr+Mo). (The solid line shows the overall average composition.) 141 0027 0022 0 017 208 203 1-98 9-65 9-15 8€5 2019 19-69 1919 • • • o ... ... Q o ° o O O O o O " u o o o • • • o o o 17 18 19 2021 22 23 24 25 26 27 230mm.-1 2 3 4 5 6 7 8 micr o probe • 9 17 18 19 20 21 2 2 23 24 25 26 27 10 II 12 13 14 15 16 Mo Ni Cr 10 12 13 14 15 16 Cr o 0 o o o _l_ o o o o o o Ni Mo _i_ O o o o _L •I L —$ — S — 58 ® to <p 5ch o cb cn ch — CVICVJ CN QQ SOU CO r-CJ r-CO O 6 Figure;36. Composition variation in conventional casting. (The solid line shows the overall average composition.) Figure 37. SEM photographs of agglomerated ferro alloy powder in area 1 in Valve No. 10. Figure 38. SEM photographs of agglomerated ferro alloy powder in area 2 in Valve No. 10. to •E 3 o o o hi • t' »• # » r; : ..* •J . Parent Metal .'I . * • 1 Area I Area 2 • •• • t. • •• •» • • • • • • • • ••• • • • * • • •••••••••• #M» • *• •••••••• »»W • •••••• • ••• M *• /K /f\ /K /N /N /K • • ••«•••••• A - . . Mo Cr Cr Fe Fe Ni Ni Mo X-ray Energy (K eV) Figure 39. EDXA plots of agglomerated ferro alloy powder in areas 1 and 2 and the parent metal (Valve No. 10). 4^ Areas 18 2 : o u o Fe-Cr Powder A * * • • • /N /K /N /N Mg Al Si S Co Ca Ti X-ray Energy (KeV) Cr Mn Fe Fe cn Figure 40 (a). EDXA plots of inclusions in areas 1 and 2 and Fe-Cr-powder. Areas I 8 2 • -.v V*.-(A C 3 O o o Fe-Cr Powder v..V.- ••„.••.•. ^ * • * *M • A • ^ ^ ^ /K /ts. /*> Mg Al Si Ca Co Ti Cr Mn Fe Fe X-ray Energy (KeV) CD Figure 40(b). EDXA plots of inclusions in areas 1 and 2 and Fe-Cr powder. S S Co Co. Fe X-ray Energy (KeV) Figure 40(c). EDXA plots of inclusions in Fe-Mo powder. Figure 41. Macroporosity in the Centre of the CF-8M electrode. 149 Figure 43. Macrostructure of the CF-8M electrode. (5.5 x) 150 Figure 44. Inclusions in the electrode piece dropped in Valve No. 9 and electrode tip (optical photo graphs). Electrode Piece 2000 x Electrode Piece 4000 x Electrode Tip 950 x Electrode Tip 1900 x Parent Casting 4000 x Parent Casting 8000 x Figure 45. Inclusions in the electrode piece dropped in Valve No. 9, electrode tip and the parent casting (Valve No. 9). Electrode Piece » .* • • *V* •••••• *.* •»... v. c o o o Electrode Tip . .. . * 4s A ^ Al Si P Ti Cr Mn Fe X-ray Energy (KeV) Fe Ni ro Figure 46(a). EDXA plots of inclusions in the electrode piece dropped in Valve No. 9 and the electrode tip. 1 4\ tt ^ 4* 4> MPS Co Ti Cr Mn Cr Fe Fe Ni X-ray Energy (K eV) co Figure 46(b). EDXA plots of inclusions in the parent casting (Valve No. 9), 154 Figure 47. Schematic to explain the peculiar pool profile observed in Valve No. 9. 155 ure 48. Macrostructure of the observed in Valve No. peculiar pool profile 9. (5.5X) 156 I I I Small Specimen (316..CF-8M and 4340) Large Specimen (316 and CF-8M) Large Specimen (4340 D 0.25 inches 1.00 inches 0.75 inches 6 1.00 " 4.00 " 3.00 " A 1.25 " 5.00 " 3.75 " R 3/16 " 3/4 9/16 " T 3/8 2 2 Figure 49. Schematic of the tensile specimens used. 80h 70 £ 60 ca 50 3 UJ 40 i 30 ca UJ or »-CO 20 10 O Tensile Strength • Yield Strength A Elongation _L 8 10 12 14 FERRITE NUMBER 16 600 500 400 o a. 300 S| or CO 200 100 18 20 Figure .50. Variation of tensile properties with ferrite number of stainless steel castings. Figure 51. Photograph of the deformed and fractured areas of small tensile specimen from Valve No. 6. Figure 52. Photograph of the deformed and fractured areas of large tensile specimen from Valve No. 6. Figure 53. AISI 4340 ESC valve. Figure 55. Radiographs of Valve No. 3. 162 163 Figure 57. Macrostructure of Valve No. 8 (HC& etch). Figure 58. Macrostructure of Valve No. 13 {HCz etch). Figure 59. Macrostructure of Valve No. 14 (Hc£ etch) (top part etched for a longer time). Figure 61 . Dendritic structure of Valve No. (a) edge (b) mid-radius (c) centre (d) top 3. Figure 62. Sulphur prints of Valve No. 3. 169 Figure 63. Sulphur Prints of Valve No. 8. 170 Figure 64. Variation of Cr, Ni and Mo across the dendritic direction in Valve No. 8. 171 i 1 1-0 mm Figure 65. Variation of Cr, Ni and Mo across the dendritic direction in Valve No. 13. 173 microprobe 15 16 17 18 19 _7_ _8_ _9_ 10 12 _I3 14 2d 21 22 23 24 ' E E m rO 1 o » o 2 O I 0 3 » o i I 4 1 ( 3 5 • c < 3 6 » o < 1 7 • o ( 8 J < 3 9 1 J 1 < i 10 4 ( » II o 3 12 3 13 o < 14 i 0 1 1 ^ < 1 tO-r <£> t£> — tO to — to SQQlf) 00 QO fs_ 00 op ^ CM CM tO rj-odd — — — odd odd Figure 67. Composition variation in Valve No. 3. (The solid line shows the overall average composition.] 046 0-41 0-36 0-27 0- 22 017 1- 94 1-84 1-74 0-84 0-79 0-74 uouuou00u000000 Ooo0u00°o r> n O n O 16 17 18 192 0 21 22 4 23 24 25 26 27 28 29 Mo Ni Cr 290 mm. 16 17 18 19 20 21 22 12 microprobe 23 2425 26 27 28 29 E E O CD C\J Cr Ni Mo 1 o • < > 2 o • < > 3 3 • < ) 4 < ) < » ( ) ( 5 0 • < > < 6 o • < > { _7 • ( > ( _8 O • ( ) _9 O • ( ) JO 0 • ( > II C • c 12 0 • c 13 o • o 14 o • c •5 0 _. 1 . • I 1 1° 1 1 * 1 <f 0) *t t 't r-OJ p- CO -r CD f> r- co co cn — c\i CVJ ro * * o o o —- - °o 6 6Qo Figure 68. Composition variation in Valve No. 8. (The solid line shows the overall average composition.) 175 0-47 0-42 0-37 026 0- 21 016 1- 94 I 84 1-74 0-90 0-65 0-80 XS—O—O—9 O 9—O D—O 9 ©" _o o o -e- -e- -e- o o o o ~o 19 20 21 22 23 12 24 25 26 27 28 Mo Ni Cr -220 mm.-20 21 2223 S 8 10 12 13 14 15 16 17 18 microprobe 24 25 26 2728 8 10 12 13 14 15 I6. 17 18 Cr D 0 D 0 o _L _L Ni Mo T" _L o m o tori co NCJ h-§§§ III sSg 8SS Figure 69. Composition variation in Valve No. 13. (The sol id'1ine shows the overall average composi tion .) 176 050 0-45 040 027 0- 22 017 202 1- 97 1-92 0-90 0-88 0-86 ho OOO0U°OOL)0U0000 o o o o o o o o o 21122123^4125 262712 28 2930 31 3233134 MO Ni Cr •295 mm. 2112212324 25E6 27 1 15 16 UJ 18 J9J microprobe 2J28|29|30|3l|32|33|34| Cr Ni Mo C 1 o • c 2 o • c 3 o • c 4 o • c 5 o • c 6 < > < i c 7 < > < < ) 8 < ) • q 9 > 0 10 11 < 0 ) • • < 0 ) 12 ( ) * ( > 13 0 • ( ) 14 o • < > 15 o • < » 16 o • c i £ 0 • ( ) 18 o • < > 19 o • < > 20 1 ? i i i Sfi » Q CO CO CO 666 CM r- CM cn cno — —Cvl 017 022 0-27 040 045 0-50 Figure 70. Composition variation in Valve No. 14. (The solid line shows the overall average composition.) (a) (b) Figure 71. Machined AISI 4340 ESC valve 178 Figure 72. Separated surfaces along a crack in Valve No. 14. gure 73. Hardness variation in heat-treated Valve No. 13. 180 I ill i i • BOTTOM I I I I I L_ 29 30 31 32 33 HARDNESS (Rc) Figure 74. Hardness variation in heat-treated Valve No. 14. 181 (c) 640 x (d) 800 x Figure 75. Microstructure AISI 4340 valve in as-cast and heat-treated conditions; electrode in heat treated condition, (a), (b) and (c) - ESC Valve No. 8 (d) - Electrode (a) 95 x (b) 175 x (c) 800 x (d) 800 x Figure 76. Microstructure of AISI 4340 ESC Valve No. 13 in heat-treated condition, (b) - white areas - ferrite areas - grey areas - black areas - bainite (shown in (c)) - Martensite (shown in (d)) Figure 77. Fractographs of large tensile specimens of AISI 4340 from ESC Valve No. 8. Figure 78. Orientation of the charpy specimens and the notch in Valve No. 8. 185 40 30 ENERGY (Ft. Lbs.) 20 T—i—r T—r i i i * ESC Valve (Long.) • Electrode (Long.) o Electrode (Trans.) J 1 1 1 L 1 1 1 -100-80-60-40-20 0 20 40 60 80 100 TEMPERATURE (°C) 60 50 40 (Joules) 30 20 10 Figure 79. Ductile Brittle transition characteristics of Valve No. 3 and the electrode. Meat Treatment: 1 hour at 845°C - oil quench. Temper at 482°C to a hardness of 39 Rc. 187 i—i 1—l—I—i—i—i—i—i—r • Trans. —i—i i i i i i i i i i -100-80-60-40-20 0 20 40 60 80 100 TEMPERATURE (°C) Figure 81. Ductile brittle transition characteristics of Valve No. 13. Heat Treatment: 1 hour at 845°C - oil quench. Temper at 560°C to a hardness of 35 Rc. 65°C 100°C 24°C 65°C 100°C Figure 82. Optical fractographs of charpy specimens from Valve No. 8 tested at different temperatures and orientations. c) - L d) - E Figure 83. Optical fractographs of charpy from the AISI 4340 electrode, a) - Long. b) Trans. specimens 191 Figure 84. SEM fractographs of charpy specimen (a) ridge area (b) micro-cracks. 192 (c) 2100 x (d) 2000 x Figure 85. SEM fractographs of different regions of a charpy specimen. (a) Area 1 (Ridge Area), 'Low Energy Fracture1. (b) Area 2 (Base of the Ridge),1 Intermediate Energy Fracture'. (c) Area 3 (General Area), 'High Energy Fracture'. (d) Area 4 (Shear Lip Area), 'High Energy Fracture'. Figure 86. Soviet electroslag cast valve. a b fixed (upside down) (a) and moving (b) core die 1-Consumable electrode; 2-Water-cooled mold (crystallizer); 3-Slag bath; 4-Metal bath; 5-casting; 6-Die; 7-Seed charge. Figure 87. Schematic of the methods used for making hollow ESC valves.3 195 Figure 90. Macrostructure of the ESC valve with the welded insert. DO NOT COPY LEAVES 197-212. APPENDIX 1 ASME/ASTM SPECIFICATIONS 197 4SI [AMERICAN NATIONAL] iSlANDAHDl ANSI/ASTM A 351 - 77 Used in USAEC-RDT Standards Standard Specification for AUSTENITIC STEEL CASTINGS FOR HIGH-TEMPERATURE SERVICE1 This Standard is issued under the fixed designation A 351; the number immediately following the designation indicates the year of original adoption or, in the case of revision, the year of last.revision. A number in parentheses indicates the year of last rcspproval. 1. Scope 1.1 This specification* covers austenilic steel castings for valves, flanges, fittings, and oiher pressure-containing parts (Note 1) in tended for high-temper2ture and corrosive service (Note 2). NOTE 1—Carbon steel castings for pressure-con taining parts are covered by Specification A 216, and low.alloy steel castings by Specification A 217. NOTE 2—The committee formulating this specifi cation has included 17 grades of materials extensively used for the present purpose. It is not the intent that this specification should be limited to these grades. Other compositions will be considered for inclusion by the committee as the need arises. Since these grades possess varying degrees of suitability for high-temperature and corrosion-resistant service, it is the responsibility of the purchaser to determine which grade shall be furnished; due consideration being riven to the requirements of the applicable construc tion codes. 1.2 Eighteen grades of austenitic steel cast ings are included in this specification. Selection will depend on design and service conditions, mechanical properties, and the high-tempera ture and corrosion-resistant characteristics. NOTE 3—The values stated in U.S. customary units are to be regarded as the standard. 2. Applicable Documents 2.1 AS^f Standards: A 216 Specification for Carbon-Steel Cast ings Suitable for Fusion Welding for High-Temperature Service* A 217 Specification for Martensitic Stainless Steel and Alloy Steel Castings for Pres sure-Containing Parts Suitable for High-Temperature Service' A 488 Recommended Practice for Qualifi cation of Procedures and Personnel for the Welding of Steel Castings4 A 703 Specification for General Require ments Applicable to Steel Castings for Pressure-Containing Parts3 E 109 Dry Powder Magnetic Particle In spection5 E 138 Wet Magnetic Particle Inspection5 E 165 Recommended Practice for Liquid Penetrant Inspection Method5 2.2 Manufacturers Standardization Society of the Valve and Fittings Industry Standard: SP 55 Quality Standard for Steel Castings for Valves. Flanges and Fittings and Other Components (Visual Method)* 3. General Conditions for Delivery 3.1 Material furnished to this specification shall conform to the applicable requirements of Specification A 703, including the supplemen tary requirements that are indicated on the purchaser's order. 3.2 The post weld heat treatment require ments of Supplementary Requirement Sll may be specified when austenitic castings other than HK or HT are to be subjected to severe corrosive service. 'This specification is under the jurisdiction of ASTM Committee A-l on Steel, Stainless Steel and Related Alloys and is the direct responsibility of Subcommittee A01.1S on Castings. Current edition approved Oct. 28, 1977. Published December 1977. Originally published as A 351 - 52 T. Last previous edition A 351 - 76. 'For ASME Boiler and Pressure Vessel Code applica tions, see related Specification SA-351 in Section II of that code. ' Annual Book of ASTM Standards. Paris I and 2. 'Annual Book of ASTM Standards, Part 2. ' A nnuol Book of ASTM Standards. Part 11. 'Available from Manufacturers Standardization Society of the Valve and Fillings Industry, 1815 N. Fort Meyer Drive, Arlington. Va. 22209. 198 4. Ordering Information 4.1 The inquiry and order shall include or indicaf the following: 4.1.1 A description of the casting by pattern number or drawing. Dimensional tolerances shall be included on the casting drawing. 4.1.2 Grade of steel. 4.1.3 Options in the specification. 4.1.4 The supplementary requirements de sired including the standards of acceptance. 5. Process 5.1 Steel shall be made by the electric furnace process. 6. Heat Treatment 6.1 All castings shall receive a heat treat ment proper to their design and chemical composition, except the HK and HT grades which shall be furnished in the as-cast condi tion. 6.2 Grade CD4MCu shall be healed to 2050°F (1120oC) for sufficient time to heat casting uniformly to temperature, furnace cooled to 1900°F (1040°C), held for a mini mum of 15 min and quenched in water or rapidly cooled by other means so as to de velop acceptable corrosion resistance. 6.3 The austenitic grades, except Grades HK and HT. shall be solution-treated by the manufacturer. 7. Chemical Requirements 7.1 The steel shall conform to the require ments as to chemical composition prescribed in Table 1. 8. Tensile Requirements 8.1 Steel used for the castings shall con form to the requirements as to tensile proper ties prescribed in Table 2. 9. Flanges 9.1 Flanged castings that have the flanges removed by machining to make welding end A 351 castings shall not be furnished under this speci fication. 10. Quality 10.1 The surface of the casting shall be ex amined visually and shall be free of adhering sand, scale, cracks, and hot tears. Other sur face discontinuities shall meet the visual ac ceptance standards specified in the order. Visual Method SP 55 or other visual standards may be used to define acceptable surface dis continuities and finish. Unacceptable visual surface discontinuities shall be removed and their removal verified by visual examination of the resultant cavities. 10.2 When additional inspection is desired, Supplementary Requirements S5, S6, and S10 may be ordered. 10.3 The castings shall not be peened, plugged, or impregnated to stop leaks. 11. Repair by Welding 11.1 Repairs shall be made using pro cedures and welders qualified under Recom mended Practice A 488. 11.2 Weld repairs shall be inspected to the same quality standards that are used to inspect the castings. When castings are produced with Supplementary Requirement S5 specified, weld repairs on castings that have leaked on hydro static test, or on castings in which the depth of any cavity prepared for repair welding exceeds 20 % of the wall thickness or 1 in. (25 mm), whichever is smaller, or on castings in which any cavity prepared for w-elding is greater than approximately 10 in.8 (65 cm*), shall be radiographed to the same standards that are used to inspect the castings. When castings are produced with Supplementary Requirement S6 specified, weld repairs shall be inspected by liquid penetrant examination to the same stan dards that are used to inspect the castings. NOTE 4—When austenitic steel castings are to be used in services where ihey will be subject to stress corrosion, the purchaser should so indicate in his order and such castings should be solution-heat treated following all weld repairs. SUPPLEMENTARY REQUIREMENTS The following supplementary requirements shall not apply unless specified in the purchase order. A list of standardized supplementary requirements for use at the option of the purchaser is included in Specification A 703. Those which are ordinarily considered suitable for use with this specification are given below. Others enumerated in A 703 may be used with this specification upon agreement between the manufacturer and purchaser. 199 S2. Destruction Tests. 55. Radiographic Inspection. 56. Liquid Penetrant Inspection. S10. Examination of Weld Preparation. S10.1 The method of performing the mag netic particle or liquid penetrant test shall be in accordance with Method E 109, Method A 351 E 138, or Recommended Practice E 165. Sll. Post Weld Heat Treatment. SI 1.1 All austenitic castings, except Grades HK and HT. which have been subjected to weld repairs shall be given a post weld solu tion heat treatment. TAni.K. 1 Chemical Hcqiiircmrnls Klcmcnl. % (max. Hsccpt Where Range is (iiven) CI-'.', CI MA ClrR. CI RA CI'.'M. CF.1MA CI-RM CIRC CI 18 CI 110 CI 120 CK20 11K .10 IIK40 irr.m CI I0MC CN7M C04MC Carbon 0.03 0.0R 0.0.1 0.08 0.08 0.08 0.10 (1.20 0.20 0.25-0.35 0.15-0.45 0.25-0.35 0.10 0.07 0.04 Manganese 1.50 I.JO 1.50 1.50 1.50 1.50 1.50 1.50 1.50 1.50 1.50 2.00 1.50 1.50 1.00 Silicon 2.00 2.00 1.50 1.50 2.00 1.50 2.01) 2.00 1.75 1.75 1.75 2.50 1.50 1.50 l.(X) Sulfur 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.04 Phosphorus 0.040 0.040 0.040 0.040 0.040 0.040 0.010 0.040 0.040 0.O40 0.040 0.040 0.040 0.040 0.04 Chromium 17.0- 18.0- 17.0- 18.0- 18.0- 22.0- 22.0- 22.0- 23.0- 23.0- 23.0- 1.1.0- 15.0- 19.0- 24.5-21.0 21.0 21.6 21.0 21.0 26.0 26.0 26.0 27.0 27.0 27.0 17.0 18.0 22.0 26.5 Nickel 8.0- 8.0- 9.0- 9.0- 9.0- 12.0- 12.0- 12.0- 19.0- 19.0- 19.0- 33.0- 13.0- 27.5- 4.75-12.0 11.0 13.0 12.0 12.0 15.0 ..15.0 15.0 22.0 22.0 22.0 37.0 16.0 30.5 6.00 Molybdenum 2.0- 20- 0.50 1.75- 2.0- 1.75-3.0 3.0 2.25 3.0 2.25 Columbium .... ... ... ... . ' ... ... ... ... ... ... ... " ... ... Copper ... ... ... ... ... ... ... ... ... ... ... ... ... J.0- 2.75-4.0 3.25 * Grade CF8C shall have a columbium contcnl of not less lhan R limes the carbon cnnlcnl but nol over 100 %. " Grade CFI0MC shall have a columbium conlenl of no! less lhan 10 limes the carbon content bul nol over 1.20 %. TABI.F. 2 Tensile Requirement CF3 CF-3A' CFR CF-8A' CF-3M CT-3MA CI-RM CIRC CUR CI 110 CI 120 CK20 1IK30 IIK40 11130 CI--I0MC CN7M CU -4MCu Tensile .strength, min, ksi 70 77 70 77 70 •80 70 70 65 70 70 65 65 62 65 70 62 100 (MPa) (4851 (530) (4R5) (530) (485) (550) (485) (485) (450) (4R5) (4R5) (450) (4 50) (425) (450) (4R5) (425) (690) Yield strength." min, ksi (MPa) 30 35 30 35 30 37 30 30 2R 30 30 2K 35 35 28 30 25 70 (205) (240) (205) (240) (205) (255) (205) (205) (195) (205) (205) (195) (240) (240) (195) (205) (170) (485) Elongation in 2 in. or 50 35.0 35.0 35.0 35.0 30.0 30.0 30.0 30.0 30 0 30.0 30.0 30.0 10.0 10.0 15.0 20.0 35.0 16.0 mm. min. % Rcduclion of area. min. % * The properties shown are obtained by adjusting the composition within the limits shown in Table t to obtain a fcrritc-ausicnitc ratio that wilt result in the higher ultimate and yield strengths indicated. Because of the thermal instability of Grades CF3A, CF3MA. and CFKA, they arc not recommended for service at temperatures in excess of 800"F (425"C). * Determine by either 0.2 % offset method or 0.5 % cxlcnsion-undc'r-load mclhod. The American Society for Testing and Materials takes no position respecting the validity of any patent rights asserted in connection with any item mentioned in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk of infringement of such rights, is entirely their own responsibility. 201 (AMERICAN NATIONAL") (STANDARD! ANSI/ASTM A 182-77a .Endorsed by Manufacturers Standardization Society of the Valve and Finings Industry Used in USAEC-RDT standards Standard Specification for FORGED OR ROLLED ALLOY-STEEL PIPE FLANGES, FORGED FITTINGS, AND VALVES AND PARTS FOR HIGH-TEMPERATURE SERVICE1 This Standard is issued under the fixed designation A 182; the number immediately following the designation indicates the year of original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of iast reapprovaj. • 1, Scope 1.1 This specification covers forged low alloy and stainless steel piping components for use in pressure systems. Included are flanges, fittings, valves, and similar pans to specified dimensions or to dimensional stand ards such as the ANSI specifications that are referenced in Section 2. 1.2 Other forgings for other applications may be made to this specification. 1.3 Thirty-three grades are covered includ ing sixteen ferritic or. martensitic steels and seventeen austenitic stainless steels. Selection will depend upon design and service require ments. 1.4 Supplementary requirements are pro vided for use when additional testing or in spection is desired. These shall apply only when specified individually by the purchaser in the order. NOTE 1—The values staled in U.S. customary units are to be regarded as the standard. 2. Applicable Documents 2.1 ASTM Standards: A 234 Specification for Piping Fittings of Wrought Carbon Steel and Alloy Steel for Moderate and Elevated Tempera tures5 A 275 Magnetic Particle Examination of Steel Forgings5 A 370 Mechanical Testing of Steel Prod ucts4 A 509 Definition of a Steel Forging3 E 30 Chemical Analysis of Steel, Cast Iron. Open-Hearth Iron, and Wrought Iron5 E 165 Recommended Practice for Liquid Penetrant Inspection Method6 E 353 Chemical Analysis of Stainless, Heat-Resisting, Maraging, and Other Similar, Chromium-Nickel-Iron Alloys5 E 381 Rating Macroetched Steel3-6 2.2 Manufacturers' Standardization Soci ety of the Valve and Fittings Industry Stand ard:7 SP 25 Standard Marking System for Valves, Fittings. Flanges and Unions. 2.3 ASME Boiler and Pressure Vessel Code:* Section IX Welding Qualifications SFA-5.4 Specification for Corrosion-Resist ing Chromium and Chromium-Nickel Steel Covered Welding Electrodes SFA-5.5 Specification for Low-Alloy Steel Covered Arc-Welding Electrodes 2.4 American National Standards Insti tute Standards:' B16.5 Dimensional Standards for Steel Pipe Flanges and Flanged Fittings 1 This specification is under the jurisdiction of ASTM Committee A-l on Steel. Stainless Steel and Related Al loys, and is the direct responsibility of Subcommittee A01.22 on Valves and Finings. Current edition approved Oct. 28. 1977. Published December 1977. Originally published as A 182 - 35 T. Last previous edition A 182 - 77. ' Annual Book of ASTM Standards, Pan 1. ' Annua! Book of ASTM Standards, Pan 5. ' Annual Book of ASTM Standards, Pans 1,2,3,4,5, and 10. • Annual Book of ASTM Standards, Pan 12. • Annual Book of ASTM Standards, Pan 1 ]. ' Available from Manufacturers' Standardization Soci ety of the Valve and Fittings Industry. 1815 N. Fon Myer Drive, Arlington, Va. 22209. 1 Available from American Societv of Mechanical Engi neers, 345 E. 47th St., New York. N.Y. 10017. • Available from American National Standards Insti tute, 1430 Broadway, New York, N. Y. 10018. 202 B16.ll Forged Steel Fittings, Socket Weld, and Threaded B16.10 Face-lo-Face and End-to-End Di mensions of Ferrous Valves 3. Basis of Purchase 3.1 Orders for material under this specifica tion shall include the following information, as necessary, to describe adequately the de-. sired material: 3.1.1 Name of forging, 3.1.2 ASTM specification number includ-• trig grade, 3.1.3 Size and pressure class or geometry, 3.1.4 Quantity, 3.1.5 Test report if required, and 3.1.6 Supplementary requirements, if any. 4. Manufacture , 4.1 The low-alloy ferritic steels may be made by the open-hearth, electric-furnace, or basic-oxygen process. The basic-oxygen process shall be limited to steels containing less than 2 % chromium. 4.2 The stainless steels shall be melted by one of the following processes: (a) electric-furnace (with separate degassing and refining optional); (b) vacuum-furnace; or (c) one of the former followed by vacuum or electro-slag-consumable remelting. Grade XM-27 may be produced by electron-beam melting. 4.3 A sufficient discard shall be made to secure freedom from injurious piping and un due segregation. . 4.4 The material shall be forged as close as practicable to the specified shape and size. Forged or rolled bar may be used without additional hot working for small cylindrically shaped parts within the limits defined by Specification A 234. 4.5 The finished product shall be a forging as defined by Definition A 509. 5. Heat Treatment 5.1 The ferritic grades and the martensitic grade shall be annealed, or normalized and tempered, except as permitted in 5.2. If fur nished in the normalized and tempered condi tion, the tempering temperature for Grades F 1, F 2, F 11 and F 12 shall be not less than 1150°F (620°C). The minimum tempering temperature for Grades F5. F5a, F6a Class 2, F7, F9, F21, and F22 shall be 1250°F A 162 - ' ;vVi-(667°C). Grade F6a Class 1 shall be tern-* pered at not less than 132ST (717°C); Grade F6a Class'3 at not less than 1100°F (593°C); and F6a Class 4 at not less than lOOO'F-. (538°C). 5.1.1 Grade F 6a (martensitic) Classes 1 and 2 need be tempered only, provided the tempering temperature for Class 1 is not less than 1325T (667°C) and for Class 2, not less than 1250°F(667eC).... 5.1.2 Grade F 6NM shall be furnished in the normalized and tempered condition; the tempering temperature shallbe not less than 1040°F (560eC) nor greater than 1120°F (600oC). 5.2 Liquid quenching followed by temper ing shall be permitted when agreed to by the purchaser. The same, minimum tempering temperature as specified in 5.1 shall be re quired for each grade. Parts that are liquid quenched and tempered shall be marked "QT" and shall be inspected for quench cracks by the magnetic particle method in accordance with Method A 275. 5.3 All austenitic material shall be fur nished in the heat-treated condition. The heat treatment shall consist of heating the material to a minimum temperature of 1900'F (1040°C) and quenching in water or rapidly cooling by other means, except for grades F 321H. F 347H, and F 348H, which shall be solution treated at 1925T (1050°C) min. 5.4 Heat treatment may be performed be fore machining. "; 6. Chemical Requirements 6.1 The steel shall conform to the require ments as to chemical composition for the grade ordered as listed in Table 1. For referee purposes, Methods E 30 or E 353 shall be used. -6.2 Grades to which lead, selenium, or other elements are added for the purpose of rendering the material free-machining shall not be used. 7. Cast or Heat (formerly Ladle) Analysis 7.1 An analysis of each heat of steel shall be made from samples taken preferably dur ing the pouring of the heat and the results shall conform to Table 1. 8. Product Analysis 8.1 The purchaser may make a product 203 analysis on forgings supplied to this specifica tion. Samples for analysis shall be taken from midway between the center and surface of solid forgings, midway between the inner and outer surfaces of hollow forgings, midway between the center and surface of full-size prolongations, or from broken mechanical test, specimens. The chemical composition thus determined shall conform to Table 1 with the tolerances as staled in Table 2 or 3. 9. Mechanical Requirements : 9.1 The material shall conform to the re quirements as to mechanical properties for the grade ordered as listed in Table 4. 9.2 Mechanical test specimens shall be ob tained from production forgings after heat treatment, or from separately forged test blanks prepared from the stock used to make the finished product. Such test blanks shall receive approximately the same working as the finished product. The test blanks shall be heat-treated with the finished product and shall approximate the maximum cross section of the forgings they represent. 9.3 For normalized and tempered, or quenched and tempered forgings, the central axis of the specimen shall correspond to the 'I* t plane or deeper position in the thickest section, /, of the represented forgings and the gage length shall be at least t distance from a second heat-treated surface. When section thickness or geometry interferes, the speci men shall be positioned as near as possible to the prescribed location. 9.4 For annealed ferritic and martensitic grades and also for austenitic stainless steels, the test, specimen may be taken from any convenient location, e : 9.5 Tension Tests:] •• •* - ..• 9.5.1 Ferritic and Martensitic Grades — One tension test shall be made for each heat in each heat treatment charge. 9.5.1.1 When the heat-treating cycles are the same and the furnaces (either batch or continuous type) are controlled within ± 25°F" (± 14°C) and equipped with recording pyrome ters so that complete records of heat treat ment are available, then only one tension test from each heat of each forging type (Note 2) and section size is required instead of one test from each heat in each heat-treatment charge. --i;.-. A 182 9.5.2 Austenitic Stainless Steel Grades-One tension test shall be made for each heat. NOTE 2 — "Type" in this case is used to describe the forging shape such as a flange, ell, tee, etc. .9.5.3 Testing shall be performed in accord ance with Methods A 370 using the largest feasible of the round specimens. The gage length for measuring elongation shall be four, times the diameter of the test section. 9.6 Hardness Tests: 9.6.1 Sufficient number of hardness meas urements shall be made in accordance with Methods A 370 to assure that the forgings are within the hardness limits given for each grade in Table 4. The purchaser may verify that the requirement has been met by testing at any location on the forging provided such testing does not render the forging useless. 9.6.2 When the reduced number of tension tests permitted by 9.5.1.1 is applied, addi tional hardness tests shall be made on forgings or samples as defined in 9.2. scattered throughout the load (Note 3). At least.eight samples shall be checked from each batch load and at least one check per hour shall be made from a continuous run. When the furnace batch is less than eight forgings, each forging shall be checked. If any check falls outside the prescribed limits, the entire lot of forgings shall be reheat treated and the requirements of 9.5.1 shall apply. NOTE 3—The tension test required in 9.5.1 is used to determine material capability and conform ance in addition to verifying the adequacy of the heat-treatment cycle. Additional hardness tests in accordance with 9.6.2 are required when 9.5.1.1 is applied to assure the prescribed heat-treating cycle and uniformity throughout the load. ... 10. Retreatment 10.1 If the results of the mechanical tests do not conform to the requirements specified, the manufacturer may reheat treat the forg ings and repeat the tests specified in Section 9. 11. Finish 11.1 The forgings shall be free of scale, machining burrs which might hinder fit-up, and other injurious imperfections as defined herein. The forgings shall have a workman like finish and machined surfaces (other than surfaces having special requirements) shall have a surface finish not to exceed 250 AA 204 (ariihmetic average) roughness height. 11.2 At the discretion of the inspector rep resenting the purchaser, finished forgings shall be subject to rejection if surface imper fections acceptable under 11.4 are not scat tered but appear over a large area in excess of what is considered to be a workmanlike fin ish. 11.3 Depth of Injurious Imperfections— Linear imperfections shall be explored for depth. When the depth encroaches on the minimum wall thickness of the finished forg ing, such imperfections shall be considered injurious. 11.4 Machining or Grinding Imperfections Not Classified as Injurious—Surface imper fections not classified as injurious shall be treated as follows: 11.4.1 Seams, laps, tears, or slivers not deeper than 5 % of the nominal wall thickness or Vi« in. (1.6 mm), whichever is less, need not be removed. If these imperfections are removed, they shall be removed by machin ing or grinding. 11.4.2 Mechanical marks or abrasions and pits shall be acceptable without grinding or machining provided the depth does not ex ceed the limitations set forth in 11.4.1. Imper fections that are deeper than '/it in. (1.6 mm), but which do not encroach on the minimum wall thickness of the forging shall be removed by grinding to sound metal. 11.4.3 When imperfections have been re moved by grinding or machining, the outside dimension at the point of grinding or machin ing may be reduced by the amount removed. Should it be impracticable to secure a direct measurement, the wall thickness at the point of grinding, or at an imperfection not required to be removed, shall be determined by deduct ing the amount removed by grinding from the nominal finished wall thickness of the forg ing, and the remainder shall be not less than the minimum specified or required wall thick ness. 12. Repair by Welding 12.1 Weld repairs shall be permitted (see Supplementary Requirement S7) at the dis cretion of the manufacturer with the following limitations and requirements: 12.1.1 The welding procedure and welders shall be qualified in accordance with Section A 182 IX of the ASME Boiler and Pressure Vessel Code. 12.1.2 The weld metal shall be deposited using the electrodes specified in Table 5. The electrodes shall be purchased in accordance with ASME Specifications SFA-5.4 or SFA-5.5. 12.1.3 Defects shall be completely re moved prior to welding by chipping or grind ing to sound metal as verified by magnetic particle inspection in accordance with Method A 275 for the ferritic or martensitic grades, or by liquid penetrant inspection in accordance with Recommended Practice E 165 for ferritic, martensitic, or austenitic grades. 12.1.4 After repair welding, the welded area shall be ground smooth to the original contour and shall be completely free of de fects as verified by magnetic-particle or liq-uid-peneirant inspection, as applicable. 12.1.5 The preheat, interpass tempera ture, and post-weld heat treatment require ments given in Table 5 shall be met. 12.1.6 Repair by welding shall not exceed 10 % of the surface area of the forging nor 33 >h % of the wall thickness of the finished forging or llt in. (9.5 mm), whichever is less, without prior approval of the purchaser. 12.1.7 When approval of the purchaser is obtained, the limitations set forth in 12.1.6 may be exceeded, but all other requirements of Section 12 shall apply. 13. Marking 13.1 Identification marks consisting of the manufacturer's symbol or name, designation of service rating, the specification number, the designation, F 1, F 2, etc.. showing the grade of material, and the size shall be legibly stamped on each forging or the forgings may be marked in accordance with Standard SP 25 of the Manufacturers' Standardization Soci ety of the Valve and Fittings Industry, and in such position so as not to injure the useful ness of the forging. 13.1.1 Quenched and tempered ferritic or martensitic forgings shall be stamped with the letters QT following the ASTM designation. 13.1.2 Forgings repaired by welding shall be marked with the letter "W" following the ASTM designation. 13.1.3 When test reports are required, the 205 0» markings shall consist of the manufacturer's symbol or name, the grade symbol, and such other markings as necessary to identify the part with the test report (13.1.1 and 13.1.2 shall apply). 14. Inspection 14.1 The manufacturer shall afford the pur chaser's inspector all reasonable facilities necessary to satisfy him that the material is being furnished in accordance with the pur chase order. Inspection by the purchaser shall not interfere unnecessarily with the man ufacturer's operations. All tests and inspec tions shall be made at the place of manufac ture unless otherwise agreed upon. 15. Certification 15.1 For forgings made to specified dimen sions, when agreed upon by the purchaser, and for forgings made to dimensional stand ards, the application of identification marks A 182 as required in 13.1 shall be the certification that the forgings have been furnished in ac cordance with the requirements of this specifi cation. 15.2 Test reports, when required, shall in clude certification that all requirements of this specification have been met, the results of all required tests, and 'the type of heat treatment. 16. Rejection 16.1 Each forging that develops injurious defects during shop working operations or in service shall be rejected and the manufac turer notified. 17. Rehearing 17.1 Samples representing material re jected by the purchaser shall be preserved until disposition of the claim has been agreed upon by the manufacturer and the purchaser. SUPPLEMENTARY REQUIREMENTS The following supplementary requirements shall apply only when specified by the purchaser in the inquiry, contract, and order. » 51. Macroetch Test Sl.l A sample forging shall be sectioned and etched to show flow lines and internal imperfections. The test shall be conducted according to Method E 381. Details of the test shall be agreed upon between the manu facturer arid the purchaser. 52. Product Analysis S2.1 A product analysis in accordance with Section 8 shall be made from one ran domly selected forging representing each size and type (Note 2) of forging on the order. If the analysis fails to comply, each forging shall be checked or the lot rejected. All re sults shall be reported to the purchaser. 53. Heat Identification and Tension Tests S3.1 In addition to the requirements of Sec tion 9, the heat identification shall be marked on each forging and one tensile specimen shall be obtained from a representative forg ing from each heat at a location agreed upon between the manufacturer and the purchaser. The results of the test shall comply with Ta ble 4 and shall be reported to the purchaser. 54. Magnetic Particle Examination S4.1 All accessible surfaces of the finished forging shall be examined by a magnetic-parti cle method. The method shall be in accord ance with Method A 275. Acceptance limits shall be as agreed upon between the manufac turer and purchaser. 55. Liquid Penetrant Examination S5.1 All accessible surfaces shall be exam ined by a liquid penetrant method in accord ance with Recommended Practice E 165. Ac ceptance limits shall be as agreed upon be tween the manufacturer and the purchaser. 56. Hydrostatic Testing S6.1 A hydrostatic test at a pressure agreed upon between the manufacturer and the purchaser shall be applied by the manufac turer. 206 • 4SID S7. Repair Welding S7.1 No repair welding shall be permitted without prior approval of the purchaser. If permitted, the restrictions of Section 12 shall apply-A 182 S8. Heat Treatment Details S8.l The manufacturer shall furnish a de-' tailed test report containing the information required in 15.2 and shall include all pertinent details of the heat-treating cycle given the forgings. ; TAM.F, I Chemical Requirement* Identi fication Symbol Composition. % Grade Phos-Carbon Manganese phorus. max Sulfur, max Silicon Nickel Chromium Molybdenum Colum bium plus Tanta lum Tanta-lum, max Tita nium Ferritic Steels F 1 carbon-molybdenum 0.28 max 0.60-0.90 0.045 0.045 0.15-0.35 0.44-0.65 F 2« 0.5 % chromium, 0.5 % molyb 0.21 max 0.30-0.80 0.040 0.04(1 0.10-0.60 0.50-0.81 0.44-0.65 denum F3» 4 to 6 % chromium 0.15 max 0.30-0.60 0.030 0.030 0.50 max 0.50 max 4.0-6.0 0.44-0.65 F 5a* 4 to 6 % chromium 0.25 max 0.60 max 0.040 0.0.30 0.50 max 0.50 max 4.0-6.0 0.44-0.65 F6a 13 % chromium 0.15 max 1.00 max 0.040 0.030 1.00 max 0.50 max 11.5-13.5 F6b 13 % chromium, 0.5 % 0.1S max 1.00 max 0.02 0.02 1.0 max 1.0-2.0 11.5-13.5 0.40-0.60 molybdenum F 6NM 13 % chromium. 4 % nickel 0.06 max 0.50-1.00 0.030 0.030 0.30-0.60 3.50-4.50 12.00-14.00 0.30-0.70 F7 6 to 8 % chromium 0.15 max 0.30-0.60 0.030 0.030 0.50-1.00 6.0-8.0 (1.44-0.65 F9 9 % chromium 0.15 max 0.30-0.60 0.030 0.030 0.50-1.00 8.0-10.0 0.90-1.10 F 11 1.25 % chromium, 0.5 % mo 0.10-0.20 0.30-0.80 0.040 0.040 0.50-1.00 1.00-1.50 0.44-0.65 lybdenum F 12 1 % chromium, 0.5 % molyb 0.10-0.20 0.30-0.80 0.040 0.040 0.10-0.60 0.80-1.25 0.44-0.65 denum F 21 chromium-molybdenum 0.15 max 0.30-0.60 0.040 0.040 0.50 max 2.65-3.35 0.80-1.06 F 22 chromium-molybdenum 0.15 max 0.30-0.60 0.040 0.040 0.50 max 2.00-2.50 0.87-1.13 F XM-27* 27 chromium, 1 molybdenum 0.010 max 0.40 max 0.020 0.020 0.40 max 0.50 max 25.00-27.50 0.75-1.50 F 429 15 chromium 0.12 max 1.00 max 0.040 0.030 0.75 max 0.50 max 14.0-16.0 F430 17 chromium 0.12 max 1.00 max 0.040 0.030 0.75 max 0.50 max 16.0-IR.O Other Elements Cu 0.50 max Other F.lcmcnts N 0.015 max Cu 0.20 max CO to Austenitic Steels F 304 18 chromium, S nickel 0.08 max 2.00 max F 30411 18 chromium, 8 nickel 0.04-0.10 2.00 max F304L 18 chromium, 8 nickel, low car- 0.035 max 2.00 max bon F304N' 18 chromium, 8 nickel, modi- 0.08 max 2.00 max Tied with nitrogen F.1I0 25 chromium, 20 nickel 0.15 max 2.00 max F316 18 chromium, 8 nickel, modi- 0.08 max 2.00 max Tied with molybdenum F 31AM 18 chromium, 8 nickel, modi- 0.04-0.10 2.00 max fied with molybdenum F3I6L 18 chromium, 8 nickel, modi- 0.035 max 2.00 max Tied with molybdenum, low carbon 0.040 0.030 1.00 max 8.00-11.00 IR.00-20 0.040 0.030 1.00 max 8.00-11.00 IR.00-20 0.040 0.030 1.00 max 8.00-13.00- 18.00-20. 0.030 0.030 0.75 max 8.00-10.50 18.00-20. 0.040 0.040 0.030 1.00 max 19.00-22.00 24.00-26 0.030 i.OOmax 10.00-14.00 16.00-18, 0.04 0 0.030 I.OOmax 10.00-14.00 16.00-18 0.040 0.030 I.OOmax 10.00-15.00 16.00-18. 00 . . . 00 . . . 00 ... .00 ... .00 ... ,00. 2.00-3.00 00 2.00-3.00 00 2.00-3.00 O <1 TAIiLF. i Continued Identi fication Symbol Composition. % Gnulc Carbon Phos-Manganese phorus, max SulHir. max Silicon Nickel Chromium Molybdenum Colli m-bium plus 'Ianta-lum Tantn-lum, max Tiln-nium F316N' 18 chromium, 8 nickel, modi 0.08 max 2.00 max 0.030 0.030 0.75 max 11.00-14.00 16.00-18.00 fied with molybdenum and ni trogen F 321 18 chromium, 8 nickel modified with titanium 0.08 max 2.00 max 0.030 0.030 1.00 max 9.00-12.00 17.00 min F 32111 ]R chromium, 8 nickel, modi fied with titanium 0.04-0.10 2.00 max 0.030 0.030 1.00 max 9.00-12.00 17.00 min F347 18 chromium, R nickel modified with columbium 0.08 max 2.00 max 0.030 0.030 1.00 max 9.00-13.00 17.00-20.00 F 34711 18 chromium, 8 nickel, modi fied with columbiuni 0.04-0.10 2.00 max 0.030 0.030 1.00 max 9.00-13.00 17.00-20.00 F348 IR chromium, 8 nickel modified with columbium 0.08 max 2.00 max 0.030 0.030 1.00 max 9.00-13.00 17.00-20.00 F 348H IR chromium, R nickel, modi fied with columbium 0.04-0.10 2.Of) max 0.030 0.030 1.00 max 9.00-13.00 17.00-20.00 FXM-19 22 chromium, 13 nickel, 5 man 0.06 max 4.00-6.00 0.040 0.030 1.00 max 11.50-13.50 20.50-23.50 ganese -3.00 -3.00 0.10-0.30 Other Elements N 0.20-0.40 V 0.10-0.30 F 10 20 nickel, 8 chromium t 0.10-0.20 0.50-0.80 0.030 0.030 1.00-1.40 19.00-22.00 7.00-9.00 * Grade F 2 was formerly assigned lo the I % chromium, 0.5 % molybdenum grade which is now Grade F 12. * The present grade F 5a (0.25 max carbon) previous to 1955 was assigned the identificatiop symbol F 5. Identification symbol F 5 in 1955 was assigned lo the 0.15 max carbon grade to be consistent with ASTM specifications for other products such as pipe, lulling, boiling, welding fittings, etc. c Grade F XM-27 shall have a nickel plus copper content of 0.50 max %. Product analysis tolerance over the maximum specified limit for carbon and nitrogen shall be 0.002 %. * Grades F 304N and F 3I6N shall have a nitrogen content of 0.10 to 0.16 %. ' Grade F 321 shall have a titanium content of not less than five times the carbon content and not more than 0.60%. 'Grade F 321H shall have a titanium content of not less lhan 4 times the carbon conlcnl and not more lhan 0.60 %. * Grades F 347 and F 348 shall have a columbium plus tantalum conlcnl of not less than ten limes the cnrbon content and not more lhan 1.00 %. * Grades F 34711 and F 348H shall have a columbium plus tantalum content of not less than 8 times Ihc carbon content and nor more lhan 1.00 %. 209 iSlI5 A 182 TABLE 2 Product Analysis Tolerances for Low-Alloy Steels Element Limit or Maximum of Tolerance Over Maximum Limit or Under Minimum Limit for Size Ranges Shown. *• Specified Range, % 100 in.1 (6.45 x 10* mm«). or less Over 100 to 200 in.' (1.290 x lO* mm'), incl Over 200 to 400 in.' (2.58! x 10= mm'), incl Over 400 in.' Manganese to 0.90 incl 0.03 0.04 0.05 0.06 over 0.90 to I.00 incl 0.04 0.05 0.06 0.07 Phosphorus to 0.045 incl 0.005 0.010 0.0)0 0.010 Sulfur to 0.045 incl 0.005 0.010 0.010 0.010 Silicon to 0.40 incl 0.02 0.02 0.03 0.04 over 0.40 to l.OO incl 0.05 0.06 0.06 0.07 Nickel to 0.50 0.03 0.03 0.03 0.03 Chromium to 0.90 incl 0.03 0.04 0.04 0.05 over 0.90 to MO incl 0.05 0.06 0.06 0.07 over 2.10 to 3.99 incl 0.10 0.10 0.12 ' 0.14 Molybdenum to 0.20 incl 0.01 0.01 0.02 0.03 over 0.20 to 0.40 incl 0.02 0.03 0.03 0.04 over 0.40 to LIS incl 0.03 0.04 0.05 0.06 •.Cross-seciionaJ area 210 lull? A 182 TABLE 4 Tensile and Hardness Requirements T.„,;I. c,,.„„-v Yield Strencih. Elongation in c . • , Brinell Hard-r-.A- c. _u,i tensile Mrenpin, , . , ., . r« Reduction of X. , Grade Symbol . w; ,WcL, mm. ksi (MPa) 2 in. or 50 mm. Zj' A ness Number, mm. ksi (MPa) (0.2 * offset) min,* Area, min, % max Fcrriiic Steels: F 1 70 (483) 40(276) 25.0 35.0 143-192 F 2 70(483) 40 (276) 20.0 30.0 143-192 F 5 70 (483) 90 (621) 40 (276) 20.0 35.0 143-217 F 5a 65 (44 S) " 22.0 50.0 187-248 F 6a Class 1 70 (483) 40 (276) 18 35.0 143-187 F 6a Class 2 85(586) 55 (379) IS 35.0 167-229 F6a Class 3' 110 (758) 85 (586) 18 35.0 207-302 F 6a Class 4' 130 (896) 1 10 (758) 18 35.0 263-321 F 6b 110-135 (758-930) 90 (621) 16 45.0 235-285 F 6NM 110-135(758-930) 90 (621) 15 35.0 235-285 F 7 70 (483) 40(276) 20.0 35.0 143-217 F 9 85 (586) 55(379) 20.0 40.0 179-217 F 11 70 (483) 40 (276) 40(276) 20.0 30.0 143-207 F 12 70 (483) 20.0 30.0 143-207 F 21 75 (517) 45 (310) 20.0 30.0 156-207 F 22 75 (517) 45 (310) 20.0 30.0 156-207 F XM-27 60 (414) 35 (241) 20.0 45.0 190 max F429 60(414) 35(241) 20.0 45.0 190 max F 4 30 60(414) 35 (241) 20.0 45.0 190 max \usienitic Steels: F 304 75 (517)" 30 (207) 30 50 F 304'H 75 (517f 30 (207) 30 50 F 304L 70 (483)° 25 (172) 30 50 F 504 N 80 (552) 35(241) 30" 501 F310 75 (517)- 30 (207) 30 50 F 316 75 (S17)r 30 (207) 30 50 F 316H 75 (517F 30 (207) 30 50 F 316L 70 (483^ 25 (172) 30 50 F 316N 80 (552) 35 (241) 30" 50" F 347 75 (517)" 30 (207) 30 50 F 347H 75 (517)° 30 (207) 30 50 F 348 75 (517)" 30(207) 30 50 F 34SH 75(517)" 30 (207) 30 50 F 321 75 (517f 30 (207) 30 50 F 321H 75 (517)° 30 (207) 30 50 F XM-19 100 (690) 55 (380) 35 55 F 10 80(552) 30 (207) 30 50 • For sections over 5 in. in thickness, the minimum tensile strength shall be 70 ksi (483 MPa). ' For sections over 5 in. in thickness, the minimum tensile sttength shall be 65 ksi (448 MPa). ' Longitudinal. The transverse elongation shall be 25 % in 2 in. or 50 mm, min. ' Longitudinal. The transverse reduction of area shall be 45 % min. ' No weld repair is permitted for these classes. 211 iSlfr A182 TABLE 3 Product Analysis Tolerances for Higher A Do; and Stainless Steels" Elements Limit or Maximum of Specified Range, % Tolerance Over the Maximum Limit or Under the Minimum Limit Carbon 0.030, ind over 0.030 to 0.20 ind 0.005 0.01 Manganese to 1.00, ind over 1.00 to 2.00, ind 0.03 0.04 Phosphorus to 0.040, ind 0.005 Sulfur to 0.030, ind 0.005 Silicon to 1.00, ind over 1.00 to 1.40, ind 0.05 0.10 Chromium over 4.00 to 10.00, ind 0.10 over 10.00 to 15.00, ind 0.15 over 15.00 to 20.00, ind 0.20 over 20.00 to 27.50, ind 0.25 Nickel to 1.00, ind 0.03 over 1.00 to 5.00, ind 0.07 over 5.00 to 10.00, ind . 0.10 over 10.00 to 20.00, ind 0.15 over 20.00 to 22.00, incl 0.20 Molybdenum over 0.20 to 0.60, incl 0.03 over 0.60 to 1.75. ind 0.05 over 1.75 to 3.00, ind ..10 Titanium aU ranges 0.05 Columbi urn- aU ranges 0.05 tantalum 0.02 Tantalum to 0.10, ind Cobalt 0.05 to 0.20, ind 0.01» Nitrogen to 0.16, ind 0.01 « This table does not apply to heat analysis. ' Product analysis limits for cohalt under 0.05% have not been established and the producer should be consulted for those limits. 212 # A 182 TABLE 5 Repair Welding Requirements Grade Symbol Ferritic Steels: F 1 F2 F 5 F 5a F6a F 6b F 6NM F 7 F9 F 11 F 12 F 21 F 11 F XM-27 F429 F430 Austenitic Steels: F304 F 304L F 304H F 304N F 310 F316 F 316L F 316K F 31bN F 321' F321H' F347 F 347H F34S F34SH F XM-19 F 10* Electrodes* Recommended Preheat and Inter-pass Temperature Range; 0 F CO Minimum Post Weld Heat-Treatment Temperature °F 0Q E 7018-A 1 E 801 g-B 1 E 502-15 or 16 E 502-15 or 16 E 410-15 or 16 13 % Cr, l'/j % Ni, '/.- % Mo 13 % Cr, 4 % Ni E 7 Cr-15 E 505-15 or 16 E 8018-B 2 E 8018-B 2 E9018-B3 E 9018-B 3 26 * Cr, 1 % Mo E 430-16 E 430-16 E 308-15 or 16 E 308L-15 or 16 E 308-15 or 16 E 308-15 or 16 E 310-15 or 16 E 316-15 or 16 E 3I6L-15 or 16 ~E 316-15 or 16 E 316-15 or 16 E 347-15 or 16 E 347-15 or 16 E 347-15 or 16 E 347-15 or 16 E 347-15 or 16 E 347-15 or 16 XM-J9W 200-400 (95-205) 300-600 (150-315) 400-700 (205-370) 400-700 (205-370) 400-700 (205-370) 400-700 (205-370) 300-700 (150-370) 400-700 (205-370) 400-700(205-370) 300-600(150-315) 300-600(150-315) 300-600(150-315) 300-*00(150-315) NR< 400-700 (205-370) NR NR NR NR NR NR NR NR NR NR NR NR NR NR NR NR NR NR 1150 (620) 1150(620) 1250 (677) 1250 (677) 1400 (760) 1150 (620) 1050 (565) 1250 (677) 1250 (677) 1150 (620) 1150 (620) 1250 (677) 1250 (677) NR 1400 (760) 1400 (760) 1900(1040) + WQ< 1900(1040) + WQ 1900(1040) + WQ 1900(1040) + WQ 1900(1040) + WQ 1900(1040) + WQ 1900(1040) + WQ 1900(1040) + WQ 1900(1040) + WQ 1900(1040) + WQ 1925 (1050) + WQ 1900(1040) + WQ 1925 (1050) + WQ 1900(1040) + WQ 1925 (1050) + WQ NR NR • Electrodes shall comply with ASME SFA 5.4 or SFA 5.5. • Purchaser approval required. ' NR = not required. • WQ = water quench. TntArntricanSo^ryforTts,^
Featured Collection
UBC Theses and Dissertations
Electroslag casting of valve bodies Gupta, Deepak 1982-12-31
Item Metadata
Download
- Media
- [if-you-see-this-DO-NOT-CLICK]
- UBC_1982_A7 G96.pdf [ 35.98MB ]
- Metadata
- JSON: 1.0078736.json
- JSON-LD: 1.0078736+ld.json
- RDF/XML (Pretty): 1.0078736.xml
- RDF/JSON: 1.0078736+rdf.json
- Turtle: 1.0078736+rdf-turtle.txt
- N-Triples: 1.0078736+rdf-ntriples.txt
- Original Record: 1.0078736 +original-record.json
- Full Text
- 1.0078736.txt
- Citation
- 1.0078736.ris
Full Text
Cite
Citation Scheme:
Usage Statistics
| Country | Views | Downloads |
|---|---|---|
| United States | 13 | 1 |
| China | 9 | 30 |
| Iran | 3 | 0 |
| Russia | 3 | 0 |
| India | 2 | 0 |
| France | 1 | 0 |
| Latvia | 1 | 0 |
| Hong Kong | 1 | 1 |
| City | Views | Downloads |
|---|---|---|
| Ashburn | 7 | 0 |
| Shenzhen | 6 | 30 |
| Unknown | 5 | 7 |
| Isfahan | 3 | 0 |
| Boardman | 2 | 0 |
| Kunming | 2 | 0 |
| Mountain View | 2 | 0 |
| Seattle | 1 | 0 |
| Central District | 1 | 1 |
| Chennai | 1 | 0 |
| Saint Petersburg | 1 | 0 |
| Kolkata | 1 | 0 |
| Beijing | 1 | 0 |
| {[{ mDataHeader[type] }]} | {[{ month[type] }]} | {[{ tData[type] }]} |
Share
Embed
Customize your widget with the following options, then copy and paste the code below into the HTML
of your page to embed this item in your website.
<div id="ubcOpenCollectionsWidgetDisplay">
<script id="ubcOpenCollectionsWidget"
src="{[{embed.src}]}"
data-item="{[{embed.item}]}"
data-collection="{[{embed.collection}]}"
data-metadata="{[{embed.showMetadata}]}"
data-width="{[{embed.width}]}"
async >
</script>
</div>
Our image viewer uses the IIIF 2.0 standard.
To load this item in other compatible viewers, use this url: http://iiif.library.ubc.ca/presentation/dsp.831.1-0078736/manifest