welding metalurgy 2.pdf

STD-AWS PRGWM-ENGL 1977 E

American Welding Society

The Practical n I n

~~

07842b5 0519407 87T W

Heterence W i U e to

I

Key Concepts for Weldability

COPYRIGHT 2003; American Welding Society, Inc.

Document provided by IHS Licensee=Shell Services International B.V./5924979112,User=, 08/21/2003 04:48:51 MDT Questions or comments about this message: pleasecall the Document Policy Management Group at 1-800-451-1584.

--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

~ ~

STD-AWS PRGWM-ENGL 1999 0784265 0519408 72b

THE PRACTICAL REFERENCE GUIDE

to WELDING METALLURGY- Key Concepts for Weldability

Compiled/edited/written by

Ted V. Weber Weber & Associates *

This pubiication is designed to provide information in regard to the subject matter covered. It is made available with the understanding that the publisher is not engaged in the rendering of professional advice. Reliance upon the information contained in this document should not be undertaken without an independent verification of its application for a particular use. The publisher is not responsible for loss or damage resulting from use of this publication. This document is not a consensus standard. Users should refer to the applicable standards for their partidar application.

American Welding Society 550 N.W. LeJeune Road, Miami, Florida 33126



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STDeAWS PRGWM-ENGL 1999 0784265 0519409 bb2

AUTHORS NOTES

For many, the metallurgical aspects of welding are not well understood and many of the books and technical articles dealing with the subject are sometimes difficult to master because the lay person does not have the technical background necessary to digest them. Generally, what welding personnel need is a basic understanding of the metallurgy of welding that is sufficient to aid in solving many of the day-to-day problems of fabrication or repair welding.

To that end, I have approached the subject less stringently than most, and have offered some basics that will aid the non-metallurgist in understanding why problems occur, and how to avoid them. While it is necessary to touch on the science in several areas, I have endeavored to limit it to the minimum needed for a practical understanding. I cover the effects of the various elements that make up our alloys, specifically from the weldability standpoint. The effects of cooling rates and the resulting structures are also covered from the mass effect and hardenability standpoints-a perspective I feel will be very helpful in understanding and solving many of the common welding problems.

I hope this Guide will be helpful to all, especially those non-metallurgists who have a need to avoid welding problems so often caused by overlooking the metallurgical considerations.

Ted V. Weber Hendersonville, Tennessee

Photocopy Rights

Authorization to photocopy items for internal, personal, or educational classroom use only, or the internal, personal, or educational classroom use only of specific clients, is granted by the American Welding Society (AWS) provided that the appropriate fee is paid to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, Tel: 978-750-8400; online: http: //www.copyright.com

O 1999 by the American Welding Society. All rights reserved. Printed in the United States of America.



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

~ ~

STD*AWS PRGWN-ENGL 1999 0784265 0519410 384

TABLE OF CONTENTS

Page No . Introduction ................................................................................................................................................................ 1 Definitions .................................................................................................................................................................. 1 Metal Structures ......................................................................................................................................................... 3 Metal Forms ................................................................................................................................................................ 5 Diffusion ..................................................................................................................................................................... 8 Solid Solubility ......................................................................................................................................................... 1 0 Shielding and Purging ............................................................................................................................................ 13

Phase Transformation ............................................................................................................................................. 15

Grain Size .................................................................................................................................................................. 20 Stainless Steels .......................................................................................................................................................... 21 Sensitization of Austenitic Stainless Steels .................................................................................. ....................... . Aluminum and its Alloys ....................................................................................................................................... 24 Copper and its Alioys ............................................................................................................................................. 25 Nickel and its Alloys ............................................................................................................................................... 25

Repair Welding ........................................................................................................................................................ 26

Residual Stress ......................................................................................................................................................... 13

Hardness and Hardenability .................................................................................................................................. 15 Effects of Elements ................................................................................................................................................... 20

Refractory Alloys ..................................................................................................................................................... 25

Summary ................................................................................................................................................................... 27 Selected References .................................................................................................................................................. 27 Glossary .................................................................................................................................................................... 28

iii COPYRIGHT 2003; American Welding Society, Inc.


--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STDoAWS PRGWM-ENGL 1777

Introduction

Knowledge of welding metallurgy can be beneficial to almost every aspect of fabrication, inspection, and failure analysis. Too often, problems occur repeatedly because the metallurgical aspects are not sufficiently understood (note Figure l), and as the old saying goes, When you continue the exact same practices, why should you expect different results?

While the subject of metallurgy, and its subset welding metallurgy, encompasses a very large technical base, there are several basic issues that can be studied and implemented to aid in avoiding problems associated with fabrication and repair welding. These basic issues will be discussed in simple terms and hopefully with an approach that will en- able a non-metallurgist to grasp and apply them in order to avoid common welding problems.

Since carbon and low-alloy steels are used predom- inantly in many industries, these alloys form the basis for much of this metallurgical review. An understanding of the steel basics can then lead to other alloy groups including austenitic stainless steels, copper and aluminum alloys, and the high alloys that include the nickel alloy groups. These families of alloys will also be discussed, but to a much lesser degree.

,

Definitions

A discussion of metals requires the first step to be a review of several basic definitions. Many definitions used in this guide are from Websters. A metal is defined as Any of a class of chemical elements generally characterized by ductility, malleability, luster, and conductivity of heat and electricity. Examples of metals include gold, iron, aluminum, and silver. Metals can be found in their natural elemental state, such as the case with gold and silver, or combined with other elements such as oxides, sulfides, sulfates, etc. These combined forms of metals are referred to as ores, and the elemental metal must be first ex- tracted, or separated from, the other constituents before combining them in desired alloy forms.

An alloy is defined as A metal that is a mixture of two or more metals, or of a metal and something else. The phrase something else in the definition can refer to the combinations of metals with ceramics, called cermets, or various other combinations. Some metal alloys occur naturally while others are combined in furnaces by intent to develop particular mechanical or physical properties. Examples of very common alloys include carbon steel, a mixture of primarily iron and carbon, and the austenitic stainless steels that are primarily mixtures of iron, chromium, and nickel. The man-made alloys also contain many other elements that may affect their properties; these will be discussed later.

Figure 1. Liberty ship failures from the World War II era: massive hull fractures due to a combination of poor-quality steel, less-than-adequate welding procedures, and low temperatures in the North Sea.

AWS Practical Reference Guide 1 COPYRIGHT 2003; American Welding Society, Inc.


--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

In addition to the iron-based, or ferrous alloys, there are nonferrous groups of alloys of aluminum, copper, and nickel, to name a few of the more common ones. Additionally, many other elements have their own specialized alloy groups, such as cobalt, tungsten, and molybdenum. Alloy development usually occurs due to a need for specialized properties not currently available. Thousands of alloys have been developed to date, and it continues at a rapid pace.

Metallurgy is defined as "The art or science of separating metalsfrom their ores, and preparing them for use by smelting and refining." This definition includes both "extractive metallurgy," the separation of metals from their ores, and "physical metallurgy," which prepares them for industrial use. Welding metallurgy is a further extension of physical metallurgy that specifically applies to those metallurgical considerations of the welding operation needed to develop an end product that can be used safely and economically. Figure 2 shows modern fabrication shops.

Metals are very unique materials permitting our industrial world to design and fabricate many very useful items not only for industrial use but for our personal needs as well. Metals vary widely in cost, availability, mechanical and physical properties, heat treatments, corrosion-resistance, weldability, and many other less common attributes. The mechanical, design, or metallurgical engineer has more than 20,000 different and unique alloys to select from for a particular application. Too often, the weldability aspects are not given enough consider-

ation at the start, and when welding difficulties occur, most individuals involved seem surprised. With the vast number of alloys available, and with the majority falling into the category of being readily weldable, one would think that welding problems would disappear. However, the art and science of joining a metal to itself, or joining two dissimilar metals, depends on much more than the initial selection of alloys. Welding a metal successfully and repeatedly requires a great number of welding variables to be considered. These welding variables include: O

O

O

O

O

O

O

O

O

Base metal chemistry, thickness, and heat treat condition

Filler metal chemistry, type, and electrode diameter

Welding process (choose from over 35 methods)

Flux system (if required)

Storage of filler metals and fluxes (heated storage, warm, dry, etc.)

Cleanliness of base and filler metals

Joint geometry (V-groove, U-groove, Square Butt, etc.)

Joint accessibility (one or both sides, open root or backup strip)

Welding heat input (volts, amps, travel speed)

Interpass temperature limits

Interpass cleaning

Figure 2. Modern fabrication shops utilizing welding equipment and procedures to avoid fabrication problems.

2 AWS Practical Reference Guide



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD-AUS PRGWM-ENGL 1999 = 0784265 O519413 093

Weather conditions (sun, rain, snow)

Ambient temperature (January in Alaska or July

Humidity (Gulf Coast or Arizona desert)

Preheat temperature

Cooling rate

Shielding gases (if required)

Purging gases (if required)

Post weld heat treatment (if required)

Others

The length of the list should not cause dismay. Sel- dom, if ever, do all the listed variables come into play for a particular weld project, but often, several of them must be considered jointly when developing welding procedures. While the study of scien- tific principles can assist in developing a welding procedure, experience is also an important factor in that development. Neither source of information can, or should, be ignored in solving welding problems since both can lead to a solution. The trial and error process played an important role in early welding development, and is still a very useful tool for solving day-to-day welding problems. A phrase that supports this process is One test can be worth a thousand expert opinions. Often, a welding test specimen must be prepared to confirm a welding procedure and this approach is still used in most code requirements. The compatibility of the base metal, filler metal, and welding process are still best deter- mined through actually joining the metal by welding a specimen and testing the result.

in Houston)

Metal Structures Metals have been in use for centuries, and the early metal artisans had little or no understanding of the science of metals. However, even without that sci- entific understanding, items made of metals found in archaeological sites confirm that metals have been worked into useful shapes for particular tasks for centuries. As shown in Figure 3, representing the early welding days, most metal working and welding was done by the blacksmith, and it relied heavily on trial and error and past experiences. Knowledge was passed on from generation to generation by word of mouth and on-the-job training. Only more recently, within the past 100 years or so,

Figure 3. Early welding was based on trial and error, and most of the time was done in the blacksmiths shop with carefully guarded procedures handed down within the family from one generation to the next.

have metals been studied in a manner to determine their detailed physical, mechanical, and chemical properties.

Initially, the metal workers used the trial-and-error process to develop alloys and the optimum heat treatments for making various products. This worked extremely well for many applications, and included the use of several alloys, including bronze and steel. The early artisans were very secretive with their metal-working knowledge in an effort to avoid competition. The early welding equipment was quite unsophisticated as shown in Figure 4. To- day, results of various alloy and welding research studies are published throughout the world.

In todays world, once the metals have been ex- tracted and refined by the extractive metallurgists, the physical metallurgists and welding engineers must have an extensive understanding of the metals physical and chemical properties to best develop them for end uses. Instruments to accurately measure alloy chemistry have been developed and their use has dramatically increased our knowledge of the metals chemical properties. We can now also measure the coefficients of thermal expansion, heat conductivity, electrical conductivity and all the other physical properties necessary to further our understanding of metals. Mechanical tests have been developed to determine the mechanical



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD*AWS PRGWM-ENGL 3999 W 07842b5 0539434 T 2 T

Figure 4. Early acetylene welding equipment.

properties so necessary for aiding the selection of alloys to perform a particular task. And lab instruments have been developed so the metals internal structure can be studied to increase our understanding of their behavior. The metallurgical microscope and the scanning electron microscope (SEM) have contributed significantly to our basic understanding of the metal structures.

Metals are composed of many crystals or grains, with grain boundaries between the adjoining individual grains. These grains are usually quite small and require polishing and etching of the metal surface to be seen on a microscope at high magnifica- tions. Within each grain, an ordered structure exists and is comprised of millions of unit cells, each following a natural physical order. At the junction of adjoining grains, there is a mismatch of each grains ordered structure since each grain is uniquely different, and this boundary of mismatch delineates each individual grain. Figure 5 represents a 200X- 400X magnification of a metal structure showing portions of three grains and their adjoining boundaries. Note the specific order within each grain that is unique to that grain alone.

The ordered structure of unit cells within a metal grain can be represented by Figure 6, which represents a three-dimensional metal structure using a unit cell approach. This natural order is comprised of metal atoms forming the unit cells. A unit cell is defined as The minimum number of atoms that filly describe each of the unique structure orientations. The sketched representation is only that; remember that the atoms are actually in motion with electrons encircling a core nucleus, and each element has a

Figure 5. Representation of a metal structure showing portions of three individual grains having an ordered structure within each, and the grain boundaries separating the grains, showing the orientation mismatch.

unique core makeup with varying numbers of electrons encircling the nucleus.

Metals usually fall into one of three types of unit cells or crystal structures: body-centered cubic (bcc), face-centered cubic (fcc), and hexagonal close-



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD-AWS PRGWM-ENGL 1999 0784265 0539435 9bb

A

Figure 6. Isometric representation of the ordering of an array of unit cells and one unit cell from the array.

packed (hcp). These three types are shown in Figure 7. Each gets its name from the orientation of the atoms in the unit celi; cubic refers to the atoms at the corners forming a cube, with each of its three axes identical in length. The terms body or face refer to the position of the additional atoms not at the corners. For the bcc structure, an additional atom is at the exact center of the cube. For the fcc, additional atoms are positioned on each of the 6 faces, with each face atom being shared with adjoining unit cells. The hcp structure is a bit more complicated, with 6 atoms forming each of the two end faces in a hexagonal shape, while the close- packed term refers to the additional three atoms between the two hexagonal end faces.

Metals and alloys can often exist with more than one crystal structure pattern, depending on the metals temperature and its rate of cooling to room temperature. There are various phase names attached to these different structures. Steel alloys can have unit cell orientations of bcc (ferrite phase), fcc (austenite phase), and an additional crystal structure known as body-centered tetragonal (bct) which is the structure of the martensite phase. The bct unit cell is very similar to bcc but with one axis longer than the other two. The bct unit cell is shown in Figure 8, and it forms on rapid quenching for

many carbon and low-alloy steels. A review of the iron-iron carbide diagram shown in Figure 9 shows the temperatures and carbon contents required for each of the various structures to be in stable forms. As seen in Figure 9, heating a 0.77% carbon steel above about 1333F causes it to transform quickly from bcc ferrite to fcc austenite. The reverse transformation occurs on slow cooling back to room temperature. Other carbon-percent alloys transform also, but over a temperature range rather than at a discrete temperature. The physical metallurgist takes advantage of these phase and structure trans- formations to develop quenching and tempering procedures to increase the mechanical strengths of alloys.

Several metals have the fcc structure at room temperature, and these include aluminum, copper, and silver. Metals having the hcp structure as one of their forms include titanium, zirconium, and cobalt. While many metals can transform on heating to al- ternate phases, thus allowing them to be strengthened by heat treatment, many cannot and these alloys must rely on other techniques for increasing their mechanical strength, such as work hardening or precipitation hardening.

As a quick review before continuing, metals are crystalline structures formed by atoms in ordered patterns. This ordered pattern or arrangement is known as a phase and is described by a unit cell. Metals solidify from many locations at once, initiating at the molten metals outer periphery, and growing toward the center of the molten puddle in preferred directions to form grains or crystals. The junction between individual grains is referred to as a grain boundary. The grain size will dictate the amount of grain boundary area present in a metal, which in turn determines to a certain degree the mechanical properties of the metal.

Metal Forms As noted above, different metals can have different atomic structures, and a single metal can exist in more than one atomic structure. Metals also have various additional forms referred to as cast, wrought, or forged. A common form for metals is the cast form; the metal is heated to above its melting point, poured into a mold, and allowed to solidify into the shape of the mold. If the casting mold is quite large and open-topped, considerable surface oxidation will occur, and significant metal



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD-AWS PRGWM-ENGL L999 0784265 0539436 8T2 .I

t I t I /

e- ----.IL-- -- * FACE-CENTERED CUBIC (fee)

HEXAGONAL CLOSE-PACKED

(hCP)

Figure 7. Representations of three common crystal structures, bcc, fcc, and hcp, showing position of individual atoms to form unit cells.

6 AWS Practical Reference Guide COPYRIGHT 2003; American Welding Society, Inc.


--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

~ ~~

STD-AWS PRGWM-ENGL 3999 M 078Y2b5 0539437 739 9

BOW-CENTERED l-rRAGONAL

(bet)

Figure 8. Unit cell of body-centered tetragonal (bct) structure showing position of individual atoms. Note the bct structure is quite similar to bcc except that one axis (the vertical axis in the sketch) is longer than the other two and therefore it is no longer cubic but rather tetragonal.

shrinkage occurs during solidification leading to a depression in the top center of the ingot. This depression is referred to as "pipe" by metallurgists. Often, this pipe is removed prior to forming the ingot into other shapes to remove its degraded quality from the final product.

Foundries have been forming metals into intricate cast shapes for many centuries since casting was one of the earliest methods for creating a particular shape. In fact, one alloy was used so extensively it became a common alloy family name, cast iron. The early uses of cast irons were very crude with resulting poor mechanical properties, but today's usage can be very refined using many of the recently developed casting alloys. For years, the inexpensive, wear-resistant, intricate castings with complex internal structures needed for automobile engines were made from cast iron and this application continues. Today, however, with automobile weight a major factor in efficiency, many lighter metals, including aluminum and magnesium alloys, are being used for automobile engines.

The "wrought" form of alloys refers to those forms wrought, or worked, into other shapes after being poured into the original cast ingot form. Steel plate begins its life as a cast ingot and through subse- quent reduction and rolling operations becomes worked into the desired thickness, width, length, and mechanical properties. Tubing can be wrought,

or drawn, by piercing an ingot to make a hole and drawing it through successive dies to obtain the desired diameters and wall thicknesses. Tubing can also be fabricated from thin sheet by forming it into a tubular shape followed by seam welding of its length, usually without the addition of filler metals (autogenous welding). See Figure 10. Shapes such as I-beams and channels are formed by "extrusion," another form of working the cast ingot into the desired shape. "Forging" refers to the working of cast ingots or bar stock material by forcing, or forging, them into the final desired shape, often at elevated temperatures. Forged-shapes are often more complex than those shapes achieved by rolling or extrusion. Improved mechanical properties can often be achieved during these working operations.

Additional terms applied to working an alloy include "hot working" and "cold working." Hot working is done above the metal's recrystallization temperature; cold working is done below the metal's recrystallization temperature but not at temperatures normally referred to as cold. Cold working of steel may be done at temperatures of 600- 800F or higher, and often the cold working itself increases the metal temperature significantly. The high nickel alloy heating coils shown in Figure 11 were formed well above room temperature to reduce the alloy's mechanical strength to aid the bending operation.



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD-AWS PRGWM-ENGL 1999 I 07842b5 0539438 b75

Figure 9.

CARBON, ATOMIC PERCENTAGE 2 4 6 8 10 12 14 16

I I I I ' " ' I I I I I '

6 PH

2800

2600

2400

2200

2000 5

1m 5 s 7600 a

E

u- + CEMEMTITE

=E

1400

1200

too0

300

200

CAST IRONS-

Iron-Iron carbide phase diagram.

Diffusion

It is usually well recognized that atoms in the gas- eous or liquid state can move about easily with re- spect to each other. A small quantity of a green gas released into a room quickly diffuses throughout the entire volume of the room. A drop of red food coloring placed into a bowl of water diffuses throughout the entire liquid volume, turning it

pink. Likewise, under certain conditions, atoms in the solid state can also change positions. In fact, within a metal, any atom may wander away, step by step, from its home position. These changes of atom position in a metal's solid state are also called "diffusion."

An example of metal diffusion is shown in Figure 12. If very smooth, flat bars of lead and gold are



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STDmAWS PRGWM-ENGL 1999 II 07842b5 0519419 501

Figure 10. Seamless extruded tubing is often used in critical shell and tube heat exchanger fabrication. Welded tubing, if made properly (Le., with sufficient cold work on final sizing to cause complete recrystallization on final heat treatment), can achieve the same high quality normally attributed to seamless tubing and is less expensive.

clamped tightly together and left at room temperature for several days, the two bars of metal remain attached when the clamps are removed. This at- tachment is due to the atoms of lead and gold mi- grating, or diffusing, into the other metal, forming a weak bond. This bond is quite weak, and the two metals can easily be broken apart by a sharp blow at their joint line. If the two metals temperatures are increased, the amount of solid state diffusion increases, and at a temperature above the melting point of both, complete mixing occurs.

Another example of diffusion within a metal occurs when the gas hydrogen is allowed to be in the vi- cinity of molten metal, such as a weld puddle. The most common source of hydrogen is moisture or organic material contamination on the surfaces of the parts to be welded. Many of the contaminants normally found on metals are organic compounds, such as oil or grease, and they contain hydrogen in their chemical makeup. The heat of welding will usually break down the water or organic contaminants into individual atoms, which includes the nascent hydrogen atom (H+).

The hydrogen atoms are quite small, and they can easily diffuse into the solid base metal structure. As they enter the base metal, the hydrogen atoms often

Figure 11. Bending tubing into heating coils often requires forming temperatures well above room temperature.

LEAD

GOLD

Figure 12. Sketch representing diffusion of gold and lead atoms with both in solid form.

AWS Practical Reference Guide 9



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

re-combine into the hydrogen molecule H2, a combination of two atoms of hydrogen that is larger in volume than two single atoms of hydrogen. The larger molecules often become trapped in the metal at discontinuities such as grain boundaries or inclu- sions. These hydrogen molecules, because of their larger size, can cause high stresses in the internal structure of the metal, and for metals of low ductility, can cause cracking. Hydrogen cracking is often referred to as underbead or delayed cracking.

To avoid diffusion of hydrogen and its propensity for causing cracks, the source of hydrogen must be eliminated or reduced, and the first step is to thor- oughly clean all surfaces to be welded. Another approach is to specify low-hydrogen electrodes for use on carbon and low-alloy steels. These low- hydrogen electrodes are specially formulated to keep their manufactured hydrogen content quite low, but they still require special handling to avoid moisture pickup after opening of the sealed ship- ping containers. Preheating the base metals is also effective in eliminating hydrogen pickup because hydrogen will diffuse out of most metals at temperatures above 225F. The methods noted above all aid in reducing the possibility of hydrogen cracking in those metals that are susceptible.

Another aspect of hydrogen that must be kept in mind is that when metals corrode, atomic, or nascent, hydrogen is formed at the cathode. If conditions are right, this nascent hydrogen is absorbed into the steel and can cause internal blistering or cracking. One common use of steel as a material of construction is for fabricating large spheres for hydrofluoric acid storage as shown in Figure 13. If the steel is not of the proper quality, hydrogen absorbed from the minor corrosion that occurs causes severe blistering problems. Approach the repair of blistering with caution as hydrogen is flammable.

Solid Solubility Most are also familiar with the normal solubility of solids into liquids. Adding a spoonful of salt to a glass of water and stirring will dissolve the salt. However, most are not familiar with one solid dissolving into another solid. In the example given previously with the lead and gold, the two metals were diffusing through solid solution into each other. And, returning to our example of salt and water, if additional salt is added, we find that some

Figure 13. Large 50-foot-diameter, carbon steel storage sphere for hydrofluoric acid. If hydrogen blistering occurs as shown in the plate cross section above, repairs can be very expensive.

of it will not dissolve regardless of how much we stir; it settles out on the bottom of the container.

The reason for the undissolved salt is that for the specific amount of liquid, and its present temperature, a critical solubility lirnif is reached. No amount of stirring will reduce the total amount of undic- solved salt. In order to dissolve more salt, the liquid volume would either have to be increased, or its temperature raised. For a solid dissolving into a liquid, there is a critical solubility limit depending on liquid volume and temperature. Metals behave sim- ilarly through solid solubility and diffusion, and they can dissolve into each other even when both are solid within their solubility limits.

But just like the salt and water, there are solubility limits for one solid dissolving into another and the critical solubility limit is dependent on the types of

1 O AWS Practical Reference Guide



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

~

STDmAWS PRGWM-ENGL 1999 W 07842b5 05I942I I b T D

solids and the temperature. Generally, the higher the temperature of a metal, the more dissolving of a second element will occur; metals can be combined or mixed even when both are still solid. As the metal temperatures are raised, the amount of diffusion and solubility increases and this aspect of metals aids in developing the various heat treatments given to alloys.

A very practical example of a solid dissolving into another solid is a method used for increasing the surface hardness of steel alloys. If the steel is packed into a bed of carbon particles, and then heated to a temperature of about 1600-1700F, which is well below the melting point of both the carbon and the steel, some of the carbon will diffuse (dissolve) into the surface of the steel. The depth of diffusion depends on the base alloy, but it is mini- mal, usually 50-90 mils (1 mil = .O01 in.). This depth of carbon diffusion is referred to as the case dep th arising from the term case hardening applied to the technique. This added carbon in the steels surface makes the surface much harder, and is useful for resisting wear and abrasion. This technique is also called pack carburizing. This ability of a steel to absorb carbon supports the need for cleanliness in welding since organic materials contain carbon which can be absorbed into steels during welding operations.

metal begins to solidify at its cooler outer edges first and after complete solidification has taken place, its structure is referred to as dendritic, having solidified by the growth of dendrites that occurs with solidification. A dendrite is a structural feature which reflects the complex shape taken by the liquid-solid interface during solidification. Figure 14 represents the orientation of these dendrites within individual grains and how these cast grains once again have a mismatch of the structures at the grain boundaries.

To further aid our understanding of welding metallurgy, the following scenario is given (see Figure 15). Consider what happens when an electric arc is initiated on the surface of an annealed metal plate from a bare, metallic tungsten electrode that does not melt during the operation (see Figure 15B). For this example, no shielding gas is used. Note also that the specific alloy type is not mentioned here; the following comments apply to most any metal alloy chemistry.

First, in a very small, localized area, the temperature rises immediately from room temperature to above the melting point of the metal (Figure 158). Temper- atures can quickly reach levels of 3000F or higher,

Another example of diffusion into a solid is a technique for increasing the surface hardness of steels. Exposing the steel surface to an ammonia environment at temperatures similar to that for carburizing causes the ammonia (NH,) to break down into its individual components of nitrogen and hydrogen, and the nitrogen atoms enter the surface of the steel. This technique is called nitriding. Both of these surface-hardening techniques demonstrate diffusion and solid solubility mechanisms for metals. Knowledge of diffusion and solid solubility will aid in understanding the importance of cleanliness in welding, as well as the need for proper shielding during all welding operations.

One approach that aids the understanding of welding metallurgy is to consider a weld as a very small casting, which it actually is. The base metal, and filler metal if added, are melted during the welding operation and solidify into the shape of the mold. For a weld, the mold is the unmelted base metal that forms the perimeter container into which the molten weld metal is placed. The as-cast weld

Figure 14. Representation of grain orientation and dendrite formation within each grain; note the mismatch of orientation from grain to grain.



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STDmAWS PRGWM-ENGL 1999 = 07842b5 0539422 OTb II

A.

B.

C.

D.

E.

F.

n

n

Figure 15. Creating an electric arc on a metal surface without shielding. Note the distortion of the base metal due to thermal expansion and contraction.

and as the metal temperature rises, thermal expansion of the hot metal surface occurs. Just prior to melting, the high temperature rise causes severe thermal expansion stresses (Figure 15C), often enough to cause distortion in the workpiece, and then as melting occurs, these thermal stresses are re- laxed since the liquid cannot continue to support the induced thermal stress (Figure 15D). A small pool of liquid is created from the melting of the base metal, and this very hot liquid now has a great affinity for contamination from the surrounding atmosphere.

The solubility of many gases is much greater in the liquid metal and constituents from the atmosphere can be quickly absorbed into the liquid when no inert shielding is used. Air contains approximately 21% oxygen and it quickly combines with the

molten metal to form metal oxides that usually have lower mechanical strength and ductility than the original metal. Thus, the unprotected metal liquid becomes a mixture of the base metal and the metal oxides. Vaporization of the hot liquid can also occur, further changing its chemical makeup, as portions of its more volatile elements are lost to the atmosphere.

When the arc is removed, the molten puddle begins to solidify at its outer edges as the cooler surrounding metal absorbs the heat from the melted zone (Figure 15D). The metal solidifies quickly, trapping the metal oxides in the structure. This sequence is shown in Figure 16. The metal that melted now has a cast structure as opposed to the original wrought structure of the base metal, and contains a much higher concentration of oxygen in the form of metal oxides. Mechanical testing of this oxidized cast structure would usually exhibit lower strength and ductility relative to the original properties of the base metal.

The bar, on cooling, will also distort due to shrinkage (Figure 15E), and the resulting distortion will create some level of residual stress (Figure 15F).

- - - #

A - INITIAL CRYSTAL FORMATION

B - CONTINUED SOLIDIFICATION

c - COMPLESOLIDIFICATION

Figure 16. Nucleation and solidification of moltenpetal, initiating at the colder, outer edgesand progressing inward.




--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD-AWS PRGWM-ENGL 1999 m 07842b5 0539423 T32 I

Shielding and Purging To avoid this oxidation damage while the metal is molten, protection of the heated zone is required, and this protection is referred to as "shielding." Shielding from contamination can be provided by various means, depending on the welding process being used:

!

Neutral combustion gases in oxyfuel welding (OM)

Vaporization of inorganic material coverings (coatings) on the metal electrode (SMAW)

Vaporization of inorganic material cores of the metal electrode (FCAW)

Vaporization of inorganic material flux covering the weld (SAW)

Inert gases surrounding the metal electrode and weld (GTAW and PAW)

Inert and reactive gas mixtures surrounding the filler metal and weld (GMAW)

Vacuum chamber

In addition to shielding of the weld face by the methods noted above, many alloys require the weld root or backside of the weld to be protected also during the welding operation. This is referred to as root shielding or a back purge. Trailing root shields supplying an inert gas are often used to provide this added shielding (purging) when required. For welding stainless steel and other high-alloy piping, the inside diameter of the pipe is purged for protection before the start of welding, usually with an inert gas. Nitrogen has also proven to be effective purge gas for many alloys.

The AWS D10.4 document, "Recommended Practices for Welding Austenitic Chromium-Nickel Stainless Steel Piping and Tubing," gives guidance for welding austenitic stainless steel pipe. Figure 17 shows the recommended purging times for pipe per foot of length and is based on a gas flow of 50 cfh. The nominal pipe size, in both inches and millimeters, is plotted along the horizontal axis. The appropriate purge time in minutes per 12-inch length is found by the pipe size intersection with the curved line multiplied by the total pipe length in feet. The example in Figure 17 shows that for 200 fe pipe, a purge time of 200 minutes, or minutes, is required.

Residual Stress Now, back to our example of melting without shielding (Figure 15). The solidified metal from our example has contracted on cooling, and the "residual stress" remaining in the component (as a result of thermal or mechanical action, or both) can approach yield point stress levels. (Similar stress levels would be present even if shielding occurred during the melting and solidification steps.) Stress levels greater than the material's yield point cannot occur because they will cause yielding and relax- ation of those higher stress levels, but the remaining residual stress can remain very high as noted. Fac- tors affecting the quantity of residual stress remaining in welded components include:

-l@

Welding process

Heat input

Thickness of base metal

Restraint of weldment

Metal temperature

Travel speed

Cooling rate

is important to note that all welding operations result in a remaining residual stress level. This level, as noted above, varies with several factors, but some level of residual stress always remains after a metal has been melted and solidified.

Reduction of the welding stress, and its related distortion problems, can be accomplished by several means. Often, "peening" operations are specified during welding to reduce residual stress. As each layer of weld metal is completed, a peening ham- mer is used to mechanically distort the hot weld metal and thereby reduce its residual stress. A caution on peening: if not controlled using developed peening procedures, cracking can be induced into the weld metal.

Another method of reducing residual stress is through "postweld heat treatment." The completed weldment is reheated to a suitable temperature thereby reducing its yield strength and the original residual stresses are reduced to a level near the low- ered level of the reduced yield strength. For carbon and low-alloy steels, thermal stress relief is usually in the 1100F to 1350F range, which reduces the metal's yield strength to the 5000-15 O00 psi range, well below its original 35 000-50 O00 psi range.



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD-AWS PRGWM-ENGL 3939 0784265 0539424 979 m

Pipe size, mm

Pipe size, in.

Preweld purge time for 12 in. (300 mm) of pipe at a flow rate of 50 CU ft per hr (23.5 liters per minute) To calculate the purge time for any length of pipe, multiply the value obtained from the chart by the length of pipe. Example: Find time required for purging of 200 f (60 meters) of 5-in. (127 mm) pipe. From chart, read one min per 12 in. (300 mm) of pipe x 200 ft (60 meters) = 200 minutes or 3 hours 20 minutes.

Figure 17. Approximate purging time for piping based on pipe diameter and length with a purge rate of 50 cfh.

14 AWC Practical Reference Guide COPYRIGHT 2003; American Welding Society, Inc.


--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD=AWS PRGWM-ENGL 1999 07842b5 0519425 805

Thus, welding on a steel having a room tempera- #! ture yield strength of 35 O00 psi can result in an as-

welded residual stress level of about 35 O00 psi. Thermal stress relief can reduce that residual stress to a much lower level of about 5O-7000 psi, a considerable reduction.

A third method of reducing residual stress is "vibratory" stress relief. This method induces high- frequency vibrations into the weldment and lowers its residual stress. Equipment and procedures for this approach are available from several vibratory stress relief vendors and can be quite effective in reducing welding distortion problems.

Phase Transformation To this point, we have reviewed the general effects of heating, cooling, shielding, and residual stress. Attention must now be given to the specific effects of cooling rates on the weld and adjoining base metal. A weld and its base metal are usually considered to have three components; these are shown in Figure 18. Starting at the left edge of the lower sketch, the first zone is the weld itself that became molten during the welding operation. The second zone is the adjoining heat-affected zone (HAZ) which did not melt but may have been affected metallurgically by the welding heat. The third zone is the original wrought base metal that remains unaffected by the welding operation.

The size and changes occurring in the weld and HAZ will vary with several factors including:

Alloy type

Alloy thickness

Welding process

Ambient temperatures

Preheat temperatures

Travel speed

Electrode diameter

Cooling rate

Hardness and Hardenability The hardness of an alloy is greatly affected by the last item noted above, cooling rate, and is also affected by many factors, including the hardenability

- W -& " A t -/ Figure 18. The three zones of a weldment starting at the left edge of the lower sketch; weld metal having a cast dendritic structure, the heat-affected zone (HAZ) showing a wrought structure with grain growth, and the original wrought base metal unaffected by welding heat.

of the alloy itself. Hardness is defined as "resistance to indentation," and most are familiar with its concepts and measurements. Hardenability is defined for ferrous alloys as "the property that determines the depth and distribution of hardness induced by quenching.'' While welds are seldom intentionally quenched immediately after welding, reviewing hardenability aspects can aid our understanding of welding metallurgy of ferrous alloys. Many weld and HAZ cracking problems in carbon and low- alloy steels are associated with the rapid cooling of the components, leading to formation of phases with high crack propensity. A review of the standard iron-iron carbide phase diagram noted earlier refreshes our memory of the various phases present



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD*AWS PRGWM-ENGL 1999 II 07842b5 051942b 741 =

at various temperatures. Go to Figure 9 on page 8 and note how this helpful diagram has the carbon percent plotted logarithmically to emphasize the importance of small amounts of carbon additions.

welded structures unless it is tempered by a post weld heat treatment. Moderate cooling rates (B, C, and D) result in mixtures of martensite, bainite, and pearlite. Bainite is a phase having less hardness

While this diagram shows the phases present at various steady-state temperatures, such as the temperature zones for ferrite, pearlite, and austenite, it does not show the effects of cooling rates on the phases formed. To determine cooling rate effects, one approach is to use the isothermal transformation (ITT) diagrams shown in Figures 19 and 20 for two different alloys, AISI 2335 and AISI 4340. These diagrams plot the initiating temperatures versus the cooling rates in time; time is plotted logarithmically in seconds. While helpful, these ITT curves are simpler to use when they are combined with a continuous cooling curve (CCT), resulting in the diagrams shown in Figures 21 and 22.

These combined curves (Figures 21 and 22) show what phases result from various cooling rates by following specific paths (A, B, C, etc.) from the austenite range to room temperature. A very fast cooling rate from the austenitizing temperature (A) results in the austenite transforming to martensite, a hard, brittle phase often causing cracking in

05 1 2 3 5 l 100 mo TIME. sec (LOO SCALE)

than martensite but greater toughness, and pearlite is a soft, ductile phase. As the cooling rates become even slower (E), the resulting phases are fine pearlite or coarse pearlite. As shown in the figures, different alloys have different responses to various cooling rates; some alloys form martensite only with a very rapid quench, while others may form martensite when cooled in air at a much lower cooling rate.

These curves emphasize the importance of cooling rates pertinent to the various alloys and provide useful data, but they do not show the effects of mass and alloying on the cooling rates. For this data, we refer to the hardenability curves developed for the various alloys. There are several tests that measure hardenability of the various alloys but one of the most common is the joominy End-Quench test. For this test, a one-inch diameter bar specimen with a larger flange at one end is prepared from an alloy and heated to its austenitizing temperature. The specimen is removed from the furnace and

I -" MARTENSITE 2M) -Mm

TIME. sec (LOG SCALE)

Figure 19. Isothermal transformation diagram for AISI 2335.

Figure 20. Isothermal transformation diagram for AISI 4340.



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

-um

Figure 21. Combined In and continuous cooling diagram for steel containing O.n% carbon.

Figure 22. Combined I l l and continuous cooling diagram for AISI 4340 steel.

dropped into the cooling fixture shown in Figure mining the alloy's CE, and then referring to the 23. Cooling occurs from a controlled water quench guidelines for preheat temperatures based on the of the unflanged end only. After quenching to room value of the CE. One CE formula from ASTh4 is: temperature, shallow, parallel flats are ground care-

%Mn %Cu %Ni %Cr %Mo %V C~ = %C + - +-+-+ ----- men, and hardness readings are taken along; the 6 40 20 10 50 10 fully 180 degrees apart on the shank of the speci-

flats beginning at the quen&ed end. The hardvness data are plotted versus distance from the quenched end as shown in Figures 24 and 25.

The guide for preheat using the above formula is shown in Table 1.

Studying the hardenability data suggests which alloys should be more difficult to weld and which require preheating to slow down the cooling rate to avoid the formation of martensite. Those alloys having low hardenability are usually much easier to weld without cracking problems, while those with high hardenability often must be preheated to be welded successfully.

Determining the need for preheating prior to welding to avoid cracking problems can also be achieved using a very useful Carbon Equivalent (CE) formula. There are many versions of the CE formula, from very simple to very complex, but all operate in the same manner. Using a CE formula requires the calculation of a CE for an alloy by inserting its actual elemental percentages into the CE formula, deter-

Table 1. Preheat guidelines for welding based on carbon equivalents

I Carbon Equivalent (CE) I Preheat Temperature 1 I c0.45

0.45-0.60

I >0.60 40O"F-70O0F

For the alloy examples noted in Figure 25, actual chemistries are listed in Table 2 with the calculated carbon equivalents.

Using the preheat guidelines in Table 1, preheat is suggested for all but the 1021 steel. Fortunately,



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD*AWS PRGWM-ENGL 1999 0784265 0519428 514 LI

1021

1050

STANDARD METHOD FOR JOMINY HARDENABILITY TEST

SEE ASTM A 255 OR !ME 5406 FOR COMPLETE DETAILS OF SPECIMEN AND PROCEDURE

JOMINY TEST SPECIMEN SPECIMEN IS MACHINED WR OF STEEL TO BE TESTED 25.4 mm (1 IN.) DIAMETER ROUND By 102 mm (4 IN.) LONG WITH FLANGE AT UPPER END AS SHOWN (OR HOOK PROVIDED)

END-QUENCHING.

HEATING

TURE SELECTED FOR AUSTENITIZING AND SOAKED ONLY LONG ENOUGH TO ACHIEVE TEMPERATURE UNIFORMITY

B W O M ROUND FACE OF SPECIMEN TO BE PROTECTED AGAINST FORMATION OF HEAVY OXIDE SCALE THAT WOULD INTERFERE WITH HEAT TRANSFER DURING QUENCHING.

END-OUENCHING APPARATUS CONSISTS OF SQUARE-END TUBE 12.7 mm (/i IN.) ID TO PROJECT STREAM OF WATER VERTICALLY UPWARD AT RATE OF 38 LITERS (ONE GALLON) PER MINUTE. THIS FLOW CORRESPONDS TO FREE FLOW HEIGHT OF 63.5 mm (2% IN.) FROM TUBE END.

WATER TEMPERATURE SHALL BE 16-27 C (SO-SO OF).

A FIXTURE ALLOWS HEATED SPECIMEN TO BE QUICKLY PLACED IN VERTICAL POSITION IN AXIAL ALIGNMENT WITH B W O M END-FACE A DISTANCE OF 12.7 mm (1h IN.) ABOVE THE ORIFICE OF THE WATER TUBE.

COOLING RATE FROM WATER END-QUENCH AT MENTAL LOCATION ALONG THE LENGTH OF JOMINY BAR CAN BE OBTAINED FROM FIG. 24.

FOR SUSPENDING IN VERTICAL POSITION FOR ili il II I I I l Ili[ m SPECIMEN IS PLACED IN FURNACE ALREADY AT TEMPERA-

6 w

z

I

. HARDENABILITY CURVE / FOLLOWMUG QUENCHING TO RT, PARALLEL FLATS 180 APART ARE GROUND ON SPECIMEN O38 mm (0.015 in.) DEEP ALONG LENGTH OF CYLINDRICAL SURFACE TO PERMIT PRECISE HARDNESS DETERMINATION.

ROCKWELL-C HARDNESS IS MEASURED ON ONE FLAT SURFACE AT SPECIFIED INCREMENTAL LOCATIONS ALONG LENGTH OF SPECIMEN AND VALUES PLOlTED AS SHOWN IN FIGS 24 AND 25.

- 0.35 0.65

0.21 0.81 - - 0.50 0.91 - - -

Figure 23. Jominy end-quench hardenability test specimen being cooled.

4150

4340

Table 2. Selected alloy analyses by percentages, and approximate carbon equivalents.

0.51 0.89 0.87 0.18 - 0.75

0.42 0.78 0.80 0.33 1.79 0.80

I I I I I I

1 1080 1 0.80 1 0.76 1 - 1 - 1 - 1 0.93 1

many of our common steel alloys can be welded without preheating. Most of the common welding grades have carbon contents below 0.35%, and many are below 0.30%, with insufficient elemental additions to require preheating. However, if cracking becomes a problem while welding carbon and low-alloy steels, adding preheat often solves the welding problems. In addition to carbon, other elements can contribute to welding ease or difficulty, and the common ones are listed on the next page.




--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

JOMINY ENMUENCH HARDENABILITY TEST DATA

FOR AISI-CAE 8630 Ni-Cr-Mo ALLOY STEEL

TO SIMULATE WELD HEAT-AFFECTED ZONE EXPOSURE, AUSTENITIZING TEMPERATURE OF 1150 Oc (2100 OF) SHOULD BE USED.

JOMINY HARDENABILITY CURVE

JOMINY POSITION, DISTANCE FROM QUENCHED END, 1/16 in. UNITS INCHES 0.5 1 .o 1.5 2 .o 2.5 3.0

MAXIMUM VALUES

w z AVERAGE VALUES u.

ABSCISSA ORIGINALLY W

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 JOMINY POSITION, DISTANCE FROM QUENCHED END, mm

Figure 24. Standard Jominy hardenability data plot.

JOMINY HARDENABILITY CURVES 70

60

50

40

30

20

10

O O 4 8 12 16 20 24 28 32 36 40

DISTANCE FROM QUENCHED END (SIXTEENTHS)

Figure 25. Jominy hardenability data for several selected alloys.

4340 LOW-ALLOY

4150 MODIFIED

1080 CARBON STEEL

1050 CARBON STEEL


~~



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD-AWS PRGWM-ENGL L979 0784265 0539430 172

Effects of Elements Aluminum is added to steel in very small amounts as a deoxidizer. It is also a grain refiner for adding improved toughness. Steels with moderate aluminum additions are referred to as having been made to afine grain practice. Aluminum additions usually improve weldability.

Carbon is generally considered to be the most effective alloying element for increasing the strength of steel and can be present up to 2% (although most welded steels have less than 0.5%). The carbon can

temperatures. Nickel additions usually improve weldability.

Phosphorus is generally considered an undesirable impurity in steels since phosphorus decreases weldability to a great degree. It is normally controlled to very low levels, and is limited to 0.04% or less in most carbon steels. In hardened steels, it may tend to cause embrittlement. In some low-alloy, high-strength steels, phosphorus may be added in amounts up to 0.10% to improve both strength and corrosion resistance but these alloys are usually not intended for welding. -

exist either dissolved in the iron, or in a combined form such as iron carbide (Fe3C). Increased amounts of carbon increase hardness and tensile strength, as well as the response to heat treatment (hardenability). Increased amounts of carbon above about 0.35% reduce weldability.

Silicon is usually added in only small amounts (0.20%) in rolled steel when it is used as a deoxidizer. However, in steel castings, 0.35 to 1.00% is commonly present. Silicon dissolves in iron and tends to strengthen it. Weld metal usually contains approximately 0.50% silicon as a deoxidizer. Some

Chromium is a powerful alloying element in steel. It is added for two principle reasons: first, it strongly increases the hardenability of steel, and second, it markedly improves the corrosion resistance of alloys in oxidizing media. Its presence in some steels can cause excessive hardness and cracking in, and adjacent to, the weld. Stainless steels contain chromium in amounts exceeding 12%.

Manganese is added to steels in amounts of at least 0.30% because it acts in a threefold manner:

It assists in deoxidizing the steel.

It prevents the formation of iron sulfide inclu- sions by preferentially forming manganese sulfides that are less detrimental.

It promotes greater strength by increasing the hardenability of the steel. Amounts up to 1.5% are found in carbon steels. Greater amounts may cause welding difficulties.

Molybdenum is a strong carbide-former and is usually present in alloy steels in amounts less than 1.0%. It is added to increase hardenability and elevated temperature strength, and can reduce weldability slightly. It is added to the austenitic stainless steels to improve pitting corrosion resistance.

Nickel is added to steels to increase their hardenability and corrosion resistance. It performs well in this function because it usually improves the toughness and ductility of the steel, even with the increased strength and hardness it brings. Nickel is frequently used to improve steel's toughness at low

f;lier metals may contain up to 1% to provide en- hanced cleaning and deoxidizing for welding on contaminated surfaces. When these filler metals are used for welding of clean surfaces, the resulting weld metal strength will be markedly increased. The resulting decrease in ductility could present cracking problems in some welding situations.

Sulfur is usually an undesirable impurity in steel rather than an alloying element. Special effort is often made to reduce it to very low levels during steel making. In amounts exceeding 0.05% it tends to cause brittleness and reduce weldability. Alloying additions of sulfur in amounts from 0.10% to 0.30% will tend to improve the machinability of steels; such types may be referred to as "resulfurized" or "free-machining." The free-machining alloys are not intended for use where welding is required.

Vanadium additions will result in an increase in the hardenability of a steel and can decrease weldability. It is very effective in this role of increasing hardenability, so it is generally added in minute amounts. In amounts greater than 0.05%, there may be a ten- dency for the steel to become embrittled during thermal stress relief treatments.

Grain Size It has been noted that metals are made up of many individual grains. These grains can vary in size by alloy and effects of working or heat treatment condition. For some services, such as forming sheet into shapes, or for low-temperature services, a




--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STDoAWS PRGWM-ENGL 1999

ASTM Grain Size I

O0

W 0784265 0539433 O09

Number per Grain Slze, mils square inch (0.001 in.)

1 I4 20.1

small grain size is desired, having good ductility and strength. Conversely, for high-temperature services, a large grain size is desired which minimizes the total areas of the grain boundaries resulting in improved high-temperature strength. Grain size numbers are used to delineate the various grain sizes, and the number is based on the number of grains fitting into one square inch. Thus, there is an inverse relationship between the grain size number and the actual size of the individual grains (the smaller the grain size number, the larger the individual grain). Table 3 shows this relationship. Typi- cally, wrought materials for ambient and low- temperature services have specified grain sizes of 6-9 while those for high-temperature services may be larger and specified to be in the 3-5 range. Table 3 shows the variability of grain size numbers, numbers of grains per square inch, and the relative grain size in mils (0.001 in.) from ASTM "Standard E 112."

3

4

5

Table 3. ASTM grain size numbers, the number of grains observed in one square inch, and the size of grains in mils (1 mil = 0.001 in.)

4 4.9

8 3.5

16 2.6

6

7

O I 112 I 14.2

32 1.8

64 1.3

I 0.5 I 3/4 I 11.8 1 1 I 9.8

I 2 I 2 I 7.1

I 8 I 128 1 0.9 I 9 I 256 I 0.6 B,

2048

4096 0.16

I 14 I 8192 I 0.12

To this point, our discussion has focused on steel and steel alloys. Other alloy groups will now be reviewed briefly to develop a basis for welding these common alloys.

Stainless Steels The word "stainless" is a bit of a misnomer when applied to the classes of metals referred to as stainless steels, since it usually means they resist corrosion. However, in severe corrosive environments, many of the stainless steels corrode at very high rates. The stainless steels are defined as having at least 12% chromium. There are many types of stainless steel, and their exact types or grades should be used when discussing them rather than the more general term, stainless steel.

Other corrosion-resisting grades are included in a separate group referred to as the "high-alloy" category. Often, these have high percentages of nickel and include the proprietary grades of Hastellop, Inconel@, and others, and these will be covered later.

The five main classes of stainless steels are ferritic, martensitic, austenitic, precipitation hardening, and the duplex grades. The first three categories refer to the stable room-temperature phase found in each class. The fourth one, often called PH stainless steels, refers to the method of hardening them by an aging heat treatment, a precipitation hardening mechanism, as opposed to the quenching and tempering mechanism known as "transformation hardening" used for carbon and low-alloy steels. The last type, the duplex grades, consists of a family of alloys that are approximately half ferrite and half austenite at room temperature with improved resistance to chloride stress corrosion cracking.

The stable room-temperature phase found in stainless steels depends on the chemistry of the steel, and some stainless steels may contain a combination of the different phases. The more common stainless steels are the austenitic grades, which are identified by the 200 and 300 series grades; 304 and 316 stainless steels are austenitic grades. Type 416 steel is a martensitic grade, and 430 is a ferritic grade. One of the common PH stainless steels is a 17-4 PH grade. A common duplex grade is 2205.

As might be expected, the weldability of these grades varies significantly. The austenitic grades are very weldable with today's available filler metal compositions. These grades can be subject to hot short cracking, which occurs when the metal is very hot. Controlling the composition of the base and filler metals to promote the formation of a sufficient quantity of a delta ferrite phase can usually elimi- nate the hot short cracking problem.

AWC Practical Reference Guide 21 COPYRIGHT 2003; American Welding Society, Inc.


--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD*AWS PRGWM-ENGL 1999 E 07842b5 0519432 T 4 5

Typically, cracking in the 300 series stainless steels Nickel Equivalent = %Ni + 30 (!OC) + 0.5 (YO-) wii be avoided by selecting filler metals with a delta ferrite percent of &lo%. This percentage is related to the ferrite number and ferrite percent can

Chromium Equivalent = %Cr + %Mo + 1.5 (%Si) + 0.5 (YoNb)

be measured with a magnetic gage. Thedelta ferrite can be measured using the magnetic gage since delta ferrite is bcc and magnetic, while the primary phase, austenite, is fcc and non-magnetic. Data have been developed to aid the selection of filler metals based on their chemistry that will result in the appropriate amounts of delta ferrite to avoid cracking. One data form for predicting ferrite percent is the Schaeffler Diagram found in Figure 26.

The Schaeffler Diagram is used to determine the approximate ferrite content based on the weld metal chemistry to calculate a nickel equivalent for the vertical axis of the chart, and a chromium equivalent for the horizontal axis using the following formulae:

Using these calculated equivalents, the percent ferrite of the weld can be approximated by plotting the two equivalents to find a distinct point on the curve that falls within a ferrite range between two ferrite percent lines. The x in Figure 26, at the black arrow, notes one such determination as an example, and indicates the ferrite will be O to 5 percent.

A modified version of the Schaeffler Diagram is referred to as the DeLong Diagram (Figure 27) and it includes the effect of nitrogen on delta ferrite formation.

The DeLong Diagram also requires calculation of chromium and nickel equivalents and plotting of the data, and it includes the effect of nitrogen on the

Figure 26. Schaeffler Diagram.



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

~ ~

STD*AWS PRGWM-ENGL 3977 E 0784265 0539433 981 =

Figure 27. DeLong Diagram.

nickel equivalent. The chromium equivalent calculation is the same as for the Schaeffler Diagram:

Nickel Equivalent = %Ni + 30 (%C) + 30 (YON)

Chromium Equivalent = %Cr + %Mo + 1.5 (%Si)

Use of this curve also permits the prediction of the ferrite content of the resulting weld to avoid hot short cracking . The ferritic stainless steels, while not used in nearly the same quantities as the austenitic stainless steels, are also considered weldable with the proper filler metals. The martensitic grades, which can be quenched and tempered to very high strength levels, are the most difficult to weld, and often require special preheating and postweld heat treatment. Procedures have been developed to weld the martensitic grades, and they must be followed carefully to avoid cracking problems and maintain the mechanical properties of the base metals. The PH and duplex stainless steels are also weldable, but again, attention must be given to the resulting changes in mechanical properties caused by welding. As

+ 0.5 (?'oh)

+ 0.5 (?om)

always, selection of filler metals is an important first step.

Sensitization of Austenitic Stainless Steels One of the common problems occurring when welding the austenitic stainless grades is that referred to as "carbide precipitation," or "sensitization." When heated into the welding temperatures, a portion of the base metal reaches temperatures in the 800-1600F range, and within this temperature range, the chromium and carbon present in the metal combine to form chromium carbides. The most severe temperature for this formation is about 1250"F, and this temperature is passed through twice on each welding operation cycle; once on heating to weld and again on cooling to room temperature.

These chromium carbides typically are found adjacent to the grain boundaries of the structure. The result of their formation is the reduction of the chromium content within the grain itself adjacent to the grain boundary, resulting in a reduction of the



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

chromium content to below that required for its maximum corrosion resistance. The final result of this chromium depletion of the grain is a reduced corrosion resistance of the grain itself due to its reduced chromium content. In certain corrosion environments, the perimeters of the grains corrode at a high rate. This is called intergranular corrosion at- tack, or IGA.

Sensitization of austenitic stainless steels during welding can be prevented by several methods. The first method involves re-heat treating the complete structure by heating to 1950-2000F and quenching rapidly in water [Solution Anneal, Water Quench (SAWQ)]. This re-heating breaks up the chromium carbides, permitting the carbon to be re-dissolved into the structure. However, this heat treatment, with its extreme water quench, can cause severe distortion of as-welded structures and is not always feasible as a solution; other solutions are often needed.

A second solution to the sensitization problem is the addition of stabilizers to the base and filler metals. The two most common examples of stabiliza- tion are the addition of titanium or niobium (columbium) to the 300 series alloys in amounts equa to 8 or 10 times the carbon content. These alloying stabilizers preferentially combine with the carbon and reduce the amount of carbon available for chromium carbide formation, maintaining the alloys chromium content and corrosion resistance. When titanium is added, we have the austenitic, stainless alloy 321; when niobium is added, we have the 347 grade.

A third solution to sensitization is the reduction of the carbon content in the base and filler metals. Ini- tially, these low-carbon austenitic stainless steels were referred to as Extra Low Carbon, or ELC for abbreviation. Today, they are referred to as L grades, and their carbon content is less than 0.030%. (The standard grades contain up to 0.08% carbon.) By reducing the carbon content in the alloy, less carbon is available to combine with the chromium, and sensitization is reduced during welding. These low- carbon grades have slightly reduced mechanical properties because of their lower carbon content, and this must be carefully considered when selecting these alloys, especially for high-temperature use where the strength reduction may be significant. Figure 28 shows the microstructure of an autogenous weld made in a low-carbon austenitic stainless steel.

Figure 28. Autogenous weld made in low-carbon austenitic stainless steel and etched to show microstructure. Note the columnar grains on either side of the weld centerline and slight grain growth in the HAZ.

Aluminum and its Alloys

Aluminum alloys are commonly used in many applications and have a very tenacious oxide film on their surface that forms very rapidly when the bare aluminum is exposed to air. These oxide films give aluminum its outstanding protection from corrosion in corrosive environments. These same oxides on the surface also interfere with the joining processes. To braze or solder these alloys, fluxes are used to break down the oxide film so the parts can be joined. When welding, the alternating-current method is also commonly used which results in a breaking down of the oxides by the current reversal of a-c welding, and reformation of the oxide film is avoided by shielding with helium or argon gas. The a-c welding method is sometimes referred to as a surface cleansing technique.

The metallurgy of aluminum and its alloys is very complex, especially regarding the great number of alloy types and heat treatments. Welding procedures must be developed incorporating the proper considerations, especially the affinity for aluminum alloys to absorb oxygen. This oxygen absorption problem often makes field welding of aluminum very difficult. The proper filler metals for most weldable grades and heat treat conditions can be found in AWS A5.10, Specification for Bare Aluminum and Aluminum Alloy Welding Electrodes and Rods.



--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STD-AWS PRGWM-ENGL 1999 = 07842b5 0519435 754

Copper and its Alloys Pure copper and many of its simpler alloys cannot by hardened by a quench and temper heat treatment as steel alloys can. These copper alloys are usually hardened and made stronger by the amount of cold work introduced when forming them into the various shapes. Copper grades are often given the terms "1 /4 hard" or "half hard" referring to the degree of cold work present. Welding on these cold-worked grades softens the cold-worked material and this must be considered before welding on work-hardened copper alloys. There is a series of copper alloys that are strengthened by aging, a treatment similar to the precipitation hardening used on the PH stainless steels. When welding on these alloys, a postweld heat treatment is usually specified to restore the alloy's original mechanical properties.

One of the major problems of welding copper and its alloys is due to their relative low melting point and very high thermal conductivity. Considerable heat must be applied to the metal to overcome its loss through conductivity, and the relatively low melting point often results in the metal melting much earlier than expected and then flowing out of the weld joint. Most copper alloys are weldable with proper technique and practice.

Nickel and its Alloys The nickel alloys are usually considered quite weldable; their ductility is very high and strength levels are moderate. Nickel dues have a propensity for ab- sorbing oxygen, and this must be considered when welding. Many of the high-nickel alloys used for severe corrosion services, such as Hastelloy@ and Inconel@, are easily welded when the appropriate filier metals and procedures are used. In fact, some of the Inconel@ filler metals are very useful for welding dissimilar alloy combinations; Inconel A@' and the Inconel 600@ filler metals are often used for this application.

Refractory Alloys The refractory metals include tantalum, titanium, and zirconium, and their cost per pound is quite high compared with other common alloys. All three

quire special consideration during welding operations. Of the three, titanium and its simpler alloys are usually the easiest to weld, but they do require careful cleaning and shielding when welding on the bench. Tantalum and zirconium are more difficult to weld, and are often welded in bagged and purged environments to exclude the ambient atmosphere, in vacuum chambers, or with special trailing shields when welded on the bench. See Figure 29.

When refractory metal welding is required, the best approach is to use fabricators who have adequate welding experience and knowledge of the materials' metallurgical aspects. All of the refractory metals can become embrittled during welding, and when embrittlement occurs, the damage cannot be reversed by heat treatment. Titanium exhibits dis- tinctive color changes when oxygen is absorbed and the alloy is damaged. This feature permits its use as a "tattle tale strip" for checking chamber purging adequacy when welding tantalum or zirconium. A small, clean strip of titanium is placed into the chamber, and when the chamber is considered adequately purged, an arc is struck onto the surface of the titanium. If the purge is adequate, the titanium molten zone will be silvery or light straw col- ored when it solidifies. If the purge is not adequate, the melted titanium surface will have various colors

(Photo courtesy of the Nooter Corporation, St. Louis, MO)

metals hive a great affinity for absoibing gases from the environment while molten, and thus re-

Figure 29. A very large, welded tantalum-lined vessel.




--```,```,`,,,`,``,```,,````,,`-`-`,,`,,`,`,,`---

STDOAWS PRGWM-ENGL L999 07842b5 05L943b b 0 m

of blues and purples on solidification, indicating severe oxidation from the environment and the need to improve the purge conditions.

Repair Welding Often, welding is the first repair method considered for damaged or corroded equipment. Most expect weld repairs to solve any and all equipment damage and rely on welding to make things right again. As most in the welding profession would agree, repairs cannot always be made as evidenced by the equipment shown in Figure 30. This vessel is not likely to lend itself to weld repair. But fortunately, most equipment damage can be weld-repaired, whether its surfacing to restore wall thickness or repairing mechanical damage. Gener- ally, when selecting a welding process for weld- repair, it is wise to choose one that involves a coating or a flux (SMAW, FCAW). The presence of these materials in the welding operation tend to be more forgiving of service contamination than those using bare wires (GTAW, GMAW). Not that the bare wire processes are excluded from repair welding, but just that the coatings and fluxes tend to have fewer cracking problems during the repairs.

There are several alternative approaches to equipment damage assessment and repair: the first should always be to consider doing nothing and

Figure 30. Not all equipment damage can be repaired; some damage requires replacement. This pressure vessel has a 4-inch wall thickness and failed due to overpressure.

determine what the expected life will be if it is put back into service as-is. This first alternative is usually overlooked because most engineers think all damage must be repaired when found. Possibly the equipment can be returned to service without repair but at a reduced pressure or temperature rating that still permits production operations. A second approach is to ask if mechanical repairs can be made, rather than welding. Figure 31 shows a

(Photos courtesy of Engineered Mechanical Services, Baton Rouge, LA.

Figure 31. Large equipment crack after penetrant test, and metal stitching repair of the crack to its right.


COPYRIGHT 2003;

welding metalurgy 2.pdf

Documents