NiDl

Nickel
Development
Institute

Guidelines for
the welded fabrication of
nickel-containing stainless steels
for corrosion resistant services

A Nickel Development Institute

Reference Book, Series No 11 007

Table of Contents
Introduction ........................................................................................................

PART I For the welder ......................................................................................

Physical properties of austenitic steels ..........................................................
Factors affecting corrosion resistance of stainless steel welds .......................
Full penetration welds ..............................................................................
Seal welding crevices ..............................................................................
Embedded iron ........................................................................................
Avoid surface oxides from welding ...........................................................
Other welding related defects ...................................................................
Welding qualifications ...................................................................................
Welder training .............................................................................................
Preparation for welding .................................................................................
Cutting and joint preparation ....................................................................
Weld joint designs ...................................................................................
Cleaning in preparation for welding ..........................................................
Oxides and other surface layers ..........................................................
Contamination elements .....................................................................
Chlorinated solvents ...........................................................................
Health hazards ...................................................................................
Fixturing, fitting and tack welding .............................................................
Fixtures and positioners ......................................................................
Backing materials ...............................................................................
Tack welding ......................................................................................
Fitting pipe joints for GTAW root welds ................................................
Purging during pipe root welding .........................................................
Welding Processes .......................................................................................
Shielded metal arc welding ......................................................................
Electrode types ..................................................................................
Other guides in SMAW .......................................................................
Electrode handling and storage ......................................................
Welding current .............................................................................
Arc starting and stopping ...............................................................
Gas tungsten arc welding .........................................................................
GTAW equipment ...............................................................................
Consumables .....................................................................................
Operator technique guides ..................................................................
Gas metal arc welding .............................................................................
Arc transfer modes .............................................................................
GMAW equipment ..............................................................................
Consumables .....................................................................................
Other welding processes ..........................................................................
Post-fabrication cleaning ...............................................................................
Surface contaminants ..............................................................................
Detection ...........................................................................................
Removal .............................................................................................
Embedded iron ........................................................................................
Detecting embedded iron ....................................................................
Removing embedded iron ...................................................................
Mechanical damage .................................................................................
Safety and welding fumes .............................................................................

1
2
2
2
2
2
3
3
3
4
4
5
5
7
7
8
9
9
9
10
10
10
11
11
12
12
13
13
13
15
15
15
16
16
17
18
18
18
19
19
19
20
20
20
20
21
21
22
22

PART II For the materials engineer ....................................................................23

Stainless steel alloys .......................................................................................23
Austenitic stainless steels ................................................................................23
Effect of welding on corrosion resistance ....................................................23
Role of weld metal ferrite ............................................................................27
Measuring weld metal ferrite ..................................................................27
Duplex stainless steels ....................................................................................28
Characteristics of duplex stainless steels ....................................................30
High temperature exposure ....................................................................30
Effect of welding on duplex stainless steels .................................................30
Nickel-enriched filler metal ....................................................................31
Heat input control ..................................................................................31
Interpass temperature control ................................................................31
Preheat .................................................................................................32
Other stainless steels ......................................................................................32
Martensitic stainless steels .........................................................................33
Ferritic stainless steels ...............................................................................33
Precipitation hardening stainless steels .......................................................34
Corrosion resistant stainless steel castings ......................................................34
Heat treatment of stainless steel ......................................................................35
Alternate 1 ............................................................................................35
Alternate 2 ............................................................................................35
Material procurement and storage guides ........................................................ 36
Surface finishes ...............................................................................................36
Purchasing guidelines ......................................................................................38
PART III For the design engineer .......................................................................39
Design for corrosion services ...........................................................................39
Tank bottoms ........................................................................................39
Tank bottom outlets................................................................................40
Bottom corner welds ..............................................................................40
Attachments and structurals ................................................................. 40
Heaters and inlets .................................................................................42
Pipe welds ............................................................................................42
Appendix A: Specifications for stainless steel for welded fabrication ........................45
Additional requirements ....................................................................................46
Bibliography ..........................................................................................................46
Acknowledgement ................................................................................................ 46
Tables
Table 1 Influence of physical properties on welding austenitic stainless steels
compared to carbon steel ................................................................ 1
Table 2 Stainless steel cutting methods ........................................................ 5
Table 3 Melting temperatures of metals and metal oxides ............................. 7
Table 4 Suggested filler metals for welding stainless steels .........................14
Table 5 Comparison of GMAW arc modes for stainless steels ......................18
Table 6 Wrought austenitic stainless steels chemical analysis .....................24
Table 7 Corrosion resistant stainless steel castings chemical analysis ..........25
Table 8 Duplex stainless steels chemical analysis .......................................29
Table 9 Duplex stainless steel filler metals typical composition ................... 32
Table 10 Suggested filler metals for welding some of the martensitic, ferritic
and precipitation hardening stainless steels ....................................33
Table 11 Stainless steel products forms ........................................................36
Table 12 Standard mechanical sheet finishes ................................................37

Figures
Figures 1-1 to 1-7 Typical weld joint designs ........................................... 6 to 7
Figure 2 Backing bar groove designs .........................................................10
Figure 3 The correct tack weld sequence ...................................................11
Figure 4 Typical pipe purging fixtures ........................................................12
Figure 5 Shielded metal arc welding ..........................................................12
Figure 6 Gas tungsten arc welding ............................................................15
Figure 7 Gas metal arc welding .................................................................17
Figure 8 Typical fabrication defects ...........................................................20
Figure 9 Crevice corrosion ........................................................................20
Figure 10 A scratch serves as an initiation site for corrosion ........................20
Figure 11 Effect of carbon control on carbide precipitation in Type 304 ........26
Figure 12 Revised constitution diagram for stainless steel weld metal ..........28
Figure 13 Typical microstructure of cold rolled, quench annealed
Alloy 2205 seamless tube ............................................................28
Figures 14-1 to 14-36 Designs for improved corrosion service ............. 39 to 44

Introduction
Nickel-containing stainless steels are
indispensable in the construction of
equipment for the process industries.
These steels are used in place of
conventional steels for properties such
as excellent corrosion resistance,
toughness at low temperatures and
good elevated-temperature properties.
The stainless steels are an excellent
choice for chemical, dairy, food, architectural, biotechnology equipment and
similar services. The wrought nickel
stainless steels widely used for corrosion services range from Type 304
(Unified Numbering System, UNS,
S30400) through the newer 6% molybdenum alloys, along with the comparable cast alloys and the duplex stainless
steels.
This publication is presented in three
sections identified as, FOR THE
WELDER (page 1), FOR THE MATERIALS ENGINEER (page 23) and FOR
THE DESIGN ENGINEER (page 39).

In the section FOR THE WELDER, it is

assumed that the welders or others
involved in welded fabrication are
familiar with the basic techniques used
in carbon steel fabrication, but have had
limited experience with nickel-containing
stainless steels. The Welder section
employs a how to do it approach for
the non-engineer but may also serve as
a reference for the welding and metallurgical engineer.
The section FOR THE MATERIALS
ENGINEER describes the various types
of stainless steels, how their metallurgical and corrosion characteristics are
affected by welding and some of the
more specialized aspects of fabrication
such as heat treating. Guidelines for
material procurement and handling are
also covered.
The section FOR THE DESIGN
ENGINEER provides a number of
design examples of how the corrosion
performance of stainless steels can be
enhanced through good design.

Part I
For the welder
Part I focuses on the fabrication and
welding of austenitic stainless steels,
Types 304, 316, 321 and 347 (UNS
S30400, S31600, S32100 and S34700)
and the more highly alloyed 4 and 6%

molybdenum stainless steels. Duplex

stainless steels which are half austenite
and half ferrite are discussed in the
section entitled For the materials
engineer.

Table I
Influence of physical properties on welding austenitic stainless steels
compared to carbon steel
Austenitic
stainless
steel
2550-2650F
(1400-1450C)

Melting point
(Type 304)

Carbon
steel
2800F
(1540C)

Remarks
Type 304 requires less heat to produce
fusion, which means faster welding for
the same heat or less heat for the same
speed.

Non-magnetic
Magnetic to over Nickel stainless steels are not subject to arc
blow.
all temperatures (1)
1300F
(705C)

Magnetic
response

Rate of heat
conductivity
(% at 212F) (100C)
(% at 1200F) (650C)

Electrical resistance
(Annealed)
(Microhm-cm, approx.)
At 68F (20C)
At 1625F (885C)

Thermal Expansion
over range indicated
-6
in./in./F x 10

-6

in./in./C x 10

(Type 304)
28%
66%

Type 304 conducts heat much more slowly

than carbon steel thus promoting sharper
heat gradients. This accelerates warping,
especially in combination with higher
expansion rates. Slower diffusion of heat
expansion through the base metal means
that weld zones remain hot longer, one
result of which may be longer dwell in the
carbide precipitation range unless excess
heat is artificially removed by chill bars, etc.

100%
100%

This is of importance in electrical fusion

methods. The higher electrical resistance of
Type 304 results in the generation of more
heat for the same current or the same heat
with lower current, as compared with carbon
steel. This, together with its low rate of heat
conductivity, accounts for the effectiveness
of resistance welding methods on Type 304.

72.0
126.0

12.5
125

9.8
(68-932F)

6.5
(68-1162F)

17.6
(20-500C)

11.7
(20-628C)

(1) Duplex stainless steels are magnetic.

Type 304 expands and contracts at a faster

rate than carbon steel, which means that
increased expansion and contraction must
be allowed for in order to control warping
and the development of thermal stresses
upon cooling. For example, more tack welds
are used for stainless steel than for carbon
steel.

For the welder

Physical properties of
austenitic stainless steels

Full penetration welds

It is well recognized that for optimum
strength, butt welds should be fullpenetration welds. In corrosion service,
any crevice resulting from lack of
penetration is also a potential site for
crevice corrosion. A typical example of
an undesirable crevice is incomplete
fusion of a pipe root pass weld such as
shown in Figure 14-31 (see page 43).
In some environments, corrosion takes
place in the crevice which, in turn, can
lead to early failure of the weld joint.

The physical properties of ordinary

carbon steels and austenitic stainless
steels are quite different and these call
for some revision of welding procedures. Physical properties, Table l,
include such items as melting point,
thermal expansion, thermal conductivity
and others that are not significantly
changed by thermal or mechanical
processing. As illustrated in Table l, the
melting point of the austenitic grades is
lower, so less heat is required to produce fusion. Their electrical resistance
is higher than that of mild steel so less
electrical current (lower heat settings) is
required for welding. These stainless
steels have a lower coefficient of thermal conductivity, which causes heat to
concentrate in a small zone adjacent to
the weld. The austenitic stainless steels
also have coefficients of thermal expansion approximately 50% greater than
mild steel, which calls for more attention
to the control of warpage and distortion.

Seal welding crevices

Crevices between two stainless steel
surfaces such as tray supports tacked
to a tank, as shown in Figure 14-16
(see page 40), also invite crevice
corrosion. Avoiding such crevices is a
design responsibility and discussed
further in the section For the Design
Engineer, as well as calling for
corrective action. However, it is helpful
for those actually making the
equipment to assist in eliminating
crevices whenever possible.

Embedded iron

Factors affecting
corrosion resistance of
stainless steel welds

When new stainless steel equipment

develops rust spots, it is nearly always
the result of embedded free iron. In
some environments, if the iron is not
removed, deep attack in the form of
pitting corrosion may take place. In less
extreme environments, the iron rust
may act as a contaminant affecting
product purity, or present an unsightly
rusty appearance to a surface that
should be clean and bright.
Free iron is most often embedded in
stainless steels during welding or
forming operations. Some cardinal
fabrication rules to follow in avoiding
free iron are:

Before discussing welding guidelines, it

is useful to describe the types of welds
and stainless steel surfaces which will
give the best performance in corrosive
environments. These are factors that
the welders or others on the shop floor
control, rather than alloy selection,
which is usually made by the end user
or Materials Engineer. The manufacture
of corrosion resistant equipment that will
give superior service should be viewed
as a joint effort of selecting the correct
alloy and then employing the proper
welding and fabrication practices. Both
elements are essential.

DO NOT bring iron or steel surfaces

into intimate contact with stainless
steel. The contact could come from
lifting tools, steel tables or storage
racks, to name a few.
2

For the welder

When, after normal precautions are
taken and there are still surface
oxides, they can be removed by acid
pickling, glass-bead blasting or one of
the other methods discussed in PostFabrication Cleaning.

DO NOT use tools such as abrasive

disks or wheels that have been previously used on ordinary iron or steel and
could have iron embedded.
Use only stainless steel wire brushes
that have never been used on carbon
steel. Never use brushes with carbon
steel wire.

Other welding related defects

Three other welding related defects
and their removal procedure are listed
below.

DO NOT leave stainless steel sheets

or plates on the floor exposed to traffic.
Sheet and plate are best stored in the
vertical position.

Arc strikes on the parent material

damage stainless steel's protective film
and create crevice-like imperfections.
Weld stop points may create pinpoint
defects in the weld metal. Both imperfections should be removed by light
grinding with clean fine grit abrasive
tools.

Locate stainless steel fabrication

away from carbon steel fabrication, if at
all possible, to avoid iron contamination
from steel grinding, cutting and blasting
operations.
The detection of free iron and removal
methods are discussed under PostFabrication Cleaning in this section.

Weld spatter creates a tiny weld

where the molten slug of metal touches
and adheres to the surface. The protective film is penetrated and tiny cervices
are created where the film is weakened
the most. Weld spatter can easily be
eliminated by applying a commercial
spatter-prevention paste to either side
of the joint to be welded. The paste and
spatter are washed off during cleanup.

Avoid surface oxides from

welding

For best corrosion resistance, the

stainless steel surface should be free of
surface oxides. The oxides may be in the
form of heat tint resulting from welding on
the reverse side or heat tint on the weld
or in the heat affected zone, HAZ. Oxides
can also develop on the root inside
diameter, ID, surface of pipe welds made
with an inadequate inert gas purge.
The oxides may vary from thin, straw
coloured, a purple colour to a black heavy
oxide. The darker the colour and heavier
the oxide, the more likely pitting corrosion
will develop, causing serious attack to the
underlying metal. It should be understood
that the oxides are harmful in corrosive
environments. Oxides normally need not
be removed when the stainless steel will
operate at high temperatures where
oxides would normally form. Heat tint
seldom leads to corrosion in atmospheric
or other mild environments but is
frequently removed for cosmetic
purposes.

Slag on some coated electrode welds

is difficult to remove completely. Small
slag particles resist cleaning and
particularly remain where there is a
slight undercut or other irregularity.
Slag particles create crevices and must
be removed by wire brushing, light
grinding or abrasive blasting with iron
free materials.

Welding qualifications
It is standard practice for fabricators of
process equipment to develop and
maintain welding procedure specifications, WPS, for the various types of
welding performed. The individual
welders and welding operators are
tested and certified by satisfactorily

For the welder

making acceptable performance qualification weldments. There are a number
of Society or Industry codes that govern
welding qualifications, but the two most
widely used in the US for corrosion
resistance equipment are:
- American Society of Mechanical
Engineers, ASME, Boiler and Pressure Vessel Code Section IX,
Welding and Brazing Qualification;
- American Welding Society, AWS,
Standard for Welding Procedure and
Performance Qualification AWS
B2.1.

P-Number
Base metal
8
Austenitic stainless steels in
Table VI from Type 304
through 347 and Alloy 254
SMO plus the similar CF cast
alloys of Table VII
10H
Duplex stainless steels includeing Alloys 255 and 2205 and
cast CD 4MCu
45
Alloys 904L and 20Cb-3 and
6% molybdenum alloys of Table VI except Alloy 254 SMO.
Not all alloys have been assigned a PNumber. Alloys without a number
require individual qualification even
though similar in composition to an alloy
already qualified. If an alloy is not listed
in the P-Number tables, the alloy manufacturer should be contacted to determine if a number has been recently
assigned by the code.

Internationally, each country typically

has its own individual codes or standards. Fortunately, there is a trend
toward the acceptance and interchange
of specifications in the interest of eliminating unmerited requalification.
Common to these codes is the identification of essential variables that establish when a new procedure qualification
test weldment is required. Essential
variables differ for each welding process
but common examples might be:
- change in base metal being welded
(P-Number);
- change in filler metal (F-Number);
- significant change in thickness being
welded;
- change in shielding gas used;
- change in welding process used.

Welder training
In complying with welding qualification
specifications such as ASME and AWS,
welders must pass a performance test.
A welder training program is not only
essential prior to taking the performance
test but also insures quality production
welding. Stainless steels are sufficiently
different in welding characteristics from
ordinary steels that the welders should
be provided training and practice time.
Once they are familiar with the stainless
steels, many welders develop a preference over regular steels. In addition to
the particular base metal and welding
process, training should also cover the
shapes to be welded such as pipe and
thin sheets or unusual welding positions.

ASME Section IX classification of PNumbers often first determines if a

separate WPS is needed. A change in a
base metal from one P-Number to
another P-Number requires
requalification. Also joints made between two base metals of different PNumbers require a separate WPS, even
though qualification tests have been
made for each of the two base metals
welded to themselves. P-Numbers are
shown below:

Preparation for welding

Stainless steels should be handled with
somewhat greater care than carbon
steels in cutting and fitting. The care
taken in preparation for welding is time

For the welder

well spent in improved weld quality and
a finished product that will give optimum
serviceability.

costs. Butt welds should be full penetration welds for corrosive services. Fillet
welds need not be full penetration as
long as both sides and ends are welded
Cutting and joint preparation to seal off voids that could collect liquid
and allow crevice corrosion.
With the exception of oxyacetylene
Fillet welding branch connections on
cutting, stainless steels can be cut by the
pipe headers leaves a large and severe
same methods used for carbon steel.
crevice on the ID. This practice invites
Oxyacetylene cutting stainless steel
crevice and microbiologically influenced
(without iron rich powder additions)
corrosion and should be prohibited for
results in the formation of refractory
stainless steel pipe fabrications in all
chromium oxides, preventing accurate,
smooth cuts. The thickness and shape of services.
The molten stainless steel weld metal
the parts being cut or prepared for weldis somewhat less fluid than carbon steel
ing largely dictates which of the methods
and depth of weld penetration is not as
shown in Table II is most appropriate.
great. To compensate, stainless steel
weld joints may have a wider bevel,
Weld joint designs
thinner land and a wider root gap. The
The weld joint designs used for
welding process also influences optistainless steels are similar to those
mum joint design. For example, spray
used for ordinary steels. The weld joint
arc, gas metal arc welding, GMAW,
design selected must produce welds of
gives much deeper penetration than
suitable strength and service performshort circuiting GMAW, so thicker lands
ance while still allowing low welding
are used with the former process.

Table II
Stainless steel cutting methods
Method
Shearing

Thickness cut
Sheet/strip, thin plate

Comments
Prepare edge exposed to
environment to remove tear
crevices.

Sawing & abrasive cutting

Wide range of thicknesses

Remove lubricant or
cutting liquid before welding or
heat treating.

Machining

Wide range shapes

Remove lubricant or
cutting liquid before welding or
heat treating.

Plasma arc cutting (PAC)

Wide range of thicknesses

Grind cut surfaces to clean metal.

Powder metal cutting with

iron-rich powder

Wide range of thicknesses

Cut less accurate than

PAC, must remove all dross,

Carbon arc cutting

Used for gouging backside

of welds and cutting
irregular shapes.

Grind cut surfaces to clean

metal.

For the welder

Typical joint designs for sheet and
plate welding are shown in Figures 1-1
through 1-5. Typical pipe joint designs
for gas tungsten arc welding, GTAW,
root welds with and without consumable inserts, are shown in Figures 1-6
and 1-7. Consumable insert rings are
widely used and are recommended for
consistent root penetration.

Figure 1-3 Typical double V joint for plate.

Figure 1-1 Typical square butt joint for

sheet.

Figure 1-4 Typical single Ujoint for plate.

Figure 1-2 Typical single V joint for

sheet and plate.
Figure 1-5 Typical double U joint for plate.
6

For the welder

Figure 1-6 Typical joint design for pipe

with consumable insert.

Cleaning in preparation for

welding
The weld area to be cleaned includes
the joint edges and two or three inches
of adjacent surfaces. Improper cleaning
can cause weld defects such as cracks,
porosity or lack of fusion. The corrosion
resistance of the weld and HAZ can be
substantially reduced if foreign material
is left on the surface before welding or a
heating operation. After cleaning, joints
should be covered unless welding will be
immediately performed.
Oxide and other surface layers
The joints to be welded should be free of
the surface oxides frequently left after

Figure 1-7 Typical joint design for pipe

welded without consumable insert.
thermal cutting. Stainless steel oxides
are comprised mainly of those of chromium and nickel. Because these oxides
melt at a much higher temperature than
the weld metal, they are not fused
during welding. Often an oxide film
becomes trapped in the solidifying weld
resulting in a defect that is difficult to
detect by radiography. This is a basic
difference from welding steel. With
steel, iron oxides melt at about the same
temperature as the weld metal. While it
is considered poor practice to weld over
a heavy steel mill scale, it does not
present the problem caused by a stainless steel oxide film. The differences
between metal and metal oxide melting
temperatures is shown in Table lll.

Table III
Melting temperatures of metals and metal oxides
Metal

Iron

Melting temperatures
F (C)

Metal
oxide

304 S/S

Fe 2O3

2850 (1565)

Fe 3O4
Nickel

2798 (1537)

Melting temperatures
F (C)

2900 (1593)

2650 (1454)

NiO

3600 (1982)

2550-2650 (1400-1454)

Cr 203

4110 (2266)

For the welder

Stainless steel wrought products
delivered by the mills are normally free
of objectionable oxides and do not
need special treatment prior to welding.
Any oxide layer would be thin and not
likely the cause of welding problems.
Very thin metals, such as strip under
0.010 in. (0.25mm) may need special
cleaning such as vapour honing since
even light oxide layers may be trapped
in small, fast solidifying welds.
Stainless steels that have been in
service often require special pre-weld
cleanup. If the alloy has been exposed
to high temperatures, the surface is
often heavily oxidized or may have a
carburized or sulphurized layer. Such
layers must be removed by grinding or
machining. Wire brushing polishes and
does not remove the tightly adhering
oxides. Stainless steel equipment that
has been in chemical service may be
contaminated by the product media. A
good example is caustic. If caustic is
left on the surface during welding, the
weld and HAZ often develops cracks.
Neutralizing caustic residue with an
acid solution is part of an effective
cleanup prior to welding. It is good
practice to give a neutralizing treatment
prior to repair welding chemical equipment. That is, neutralize acid contaminated surfaces with a mild basic solution and an alkaline contaminated
surface with a mild acidic solution. A
hot water rinse should always follow the
neutralizing treatment.

sulphur, carbon

- hydrocarbons
such as cutting
fluids, grease, oil,
waxes and
primers

sulphur,
phosphorous,
carbon

- marking crayons,
paints and temperature indicating
markers

lead, zinc,
copper

- tools such as
hammers (lead),
hold down or
backing bars
(copper), zinc rich
paint

shop dirt

- any or all of the

above

The presence of sulphur, phosphorous

and low-melting metals may cause
cracks in the weld or HAZ. Carbon or
carbonaceous materials left on the
surface during welding may be taken in
solution, resulting in a high carbon layer
which in turn lowers the corrosion
resistance in certain environments.
Cleaning to remove the above contaminants should be accomplished by
following a few guidelines, along with
common sense. Metallic contaminants
and materials not having an oil or grease
base are often best removed by
mechanical means such as abrasive
blasting or grinding. It is essential that
the blasting material or abrasive disk be
free of contaminants such as free iron. A
Contamination elements There are nitric acid treatment, followed by neua number of elements and compounds
tralization can also effectively remove
that must be removed from the surface
some low melting metals without damprior to welding. If not removed, the heat age to the stainless steel.
from welding can cause cracking, weld
Oil or grease (hydrocarbon) base
defects or reduced corrosion resistance contaminants must be removed by
of the weld or HAZ. The elements to be solvent cleaning because they are not
avoided and common sources of the
removed by water or acid rinses. Large
elements are:
weldments are usually hand cleaned by
wiping with solvent saturated cloths.
Other acceptable methods include
immersion in, swabbing with or spraying
with alkaline, emulsion, solvent or
8

For the welder

misapplication, some organizations
prohibit the use of any chlorinated
solvent across the board. Non-chlorinated solvents are preferred for cleaning
stainless steels and should always be
used for equipment and crevices.

detergent cleaners or a combination of

these; by vapour degreasing; by steam,
with or without a cleaner; or by highpressure water jetting. American Society
for Testing and Materials, ASTM, A380,
Standard Recommended Practice for
Cleaning and Descaling Stainless Steel
Parts, Equipment and Systems, is an
excellent guide for fabricators and users.
A typical procedure to remove oil or
grease includes:
- remove excess contaminant by
wiping with clean cloth;
- swab the weld area (at least 2
in.(5cm) each side of the weld) with
an organic solvent such as aliphatic
petroleums, chlorinated hydrocarbons or blends of the two. (See
cautionary remarks below.) Use only
clean solvents (uncontaminated with
acid, alkali, oil or other foreign
material) and clean cloths;
- remove all solvent by wiping with
clean, dry cloth;
- check to assure complete cleaning.
A residue on the drying cloth can
indicate incomplete cleaning. Where
size allows, either the water-break or
atomized test are effective checks.

Health hazards The term health

hazard has been defined as including
carcinogens, toxic agents, irritants,
corrosives, sensitizers and any agent
that damages the lungs, skin, eyes or
mucous membranes. Each organization
should assure that the solvents used are
not harmful to personnel or equipment.
In addition to the toxic effect, consideration must be given to venting of explosive fumes, safe disposal of spent
solutions and other related handling
practices. Compliance with state and
local regulations is obviously a requirement.
Solvents used for pre-weld cleaning
include, but are not limited to, the
following:
- non-chlorinated: toluene, methyl ethyl ketone and acetone
- chlorinated solvent: 1.1.1.
Trichloroethane
All must be handled in compliance with
regulator requirements and manufacturers' instructions.

Selecting the solvent cleaner involves

considerations more than just the
ability to remove oil and grease. Two
precautions are as follows.

Fixturing, fitting and tack

welding
Good alignment of the assembly prior
to welding can reduce welding time. It is
essential that the mating pieces to be
joined should be carefully aligned for
good quality welding. When one member is considerably thicker than the
other, for example a tank head thicker
than the shell, the head side should be
machined to a taper of 3:1 or more to
reduce stress concentrations. Joints
with varying root gap require special
adjustment by the welding operator and
may result in burn through or lack of
penetration. When the volume of identical parts is large, use of fixtures is often
economically justified.

Chlorinated solvents Many

commercial solvents contain chlorides
and are effective in cleaning machined
parts and crevice free components. The
potential problem with chlorinated
solvents is that they may remain and
concentrate in crevices and later initiate
crevice corrosion and stress corrosion
cracking, SCC. There have been unnecessary and costly SCC failures of
stainless steel heat exchangers after
cleaning with chlorinated solvents.
Cleaning of open, bold areas with
chlorinated solvents does not present a
problem, but rather than risk a
9

For the welder

most often used for backing bars.
Typical backing bar designs for use with
and without a backing gas are shown in
Figure 2. In the normal course of welding, the copper bar chills the weld to
solid metal without melting the copper.
The arc should not be misdirected to the
extent that copper is melted and incorporated into the stainless steel weld or
weld cracking can result. It is good
practice to pickle after welding to remove
traces of copper from the surface and
essential to pickle if solution annealing is
to follow welding.
Argon backing gas provides excellent
protection to the underneath side of
GTAW welds. It helps control penetration
and maintain a bright, clean under
surface. Nitrogen is also used as a
backing gas and has a price advantage
over argon. However, nitrogen should
not be introduced into the arc atmosphere which, in turn, could alter the weld
metal composition balance.
When a copper backing bar or an inert
gas backing purge is impractical, there
are commercially available tapes, pastes
and ceramic backing products. These
offer some protection from burn-through
but give little protection from oxidation,
so final cleaning by abrasive means or
acid pickling is needed after welding
when these backing materials are used.

Fixtures and positioners Fixtures

are usually designed for each particular
assembly and hold the parts together
throughout the welding operation. When
fixtures are attached to positioners,
there is a further advantage in that
welding can be done in the most convenient position. Some advantages of
using fixtures are:
- better joint match-up;
- less tacking and welding time;
- distortion from welding is minimized;
- finish assembly is made to closer
tolerance.
It is important that fixture surfaces
holding the stainless steel parts do not
introduce iron contamination. This can be
avoided by surfacing the fixture contacting surfaces with stainless steel and
using the fixtures only for stainless steel.
Backing materials A backing
material should be used in welding sheet
or plate, unless both sides of the joint
can be welded. Without a backing, the
underneath side may have erratic
penetration with crevices, voids and
excessive oxidation. Such defects
reduce weld strength and can initiate
accelerated corrosion. Copper, with its
high thermal conductivity, is the material

Tack welding Joints not held in

fixtures must be tack welded to maintain
a uniform gap and alignment along the
entire length. The tacks should be
placed in a sequence to minimize the
effect of shrinkage. In fitting two sheets,
tack welds should be placed at each
end and then the middle section as
shown in Figure 3 (A). Figure 3 (B)
shows how the sheets close up when
tack welding progresses from one end.
Tack welds in stainless steel must be
spaced considerably closer than would
be needed for ordinary carbon steel
since the higher thermal expansion of
stainless steel causes greater distortion.
A rough guide is to use about half the

Figure 2 Backing bar groove designs.

(A) Standard groove for use without a
backing gas. (B) Square-corner
groove employed with backing gas.
10

Figure 3 The correct tack weld sequence is shown in A above. When tack
welding from one end only, as shown in
B, the edges close up.
spacing between stainless tacks as used
for carbon steel when distortion is a
factor.
The length of tack welds may be as
short as 0.125 in. (3 mm), or a small spot
of weld metal for thin material to over 1
in. (254 mm) long for heavy plate
sections. More important, the shape of
the tack should not cause a defect in the
final weld. Heavy or high tacks or abrupt
starts and stops should be contour
ground. Bead shape is easier controlled
with the GTAW process, making it a
good choice for tack welding. Tack welds
to be incorporated into the final weld
must be wire brushed or ground to clean
metal. They should be inspected for
crater cracks and any cracks ground out.
Fitting pipe joints for GTAW root
welds Tack welding is important
because the tack normally becomes a
part of the root weld. Inert gas purging
prior to tacking is needed for protection
against oxidation. In tacking joints
without consumable inserts, or open root
welds as they often are called, there is a
strong tendency for the shrinkage forces

For the welder

to pull the joint closed. To maintain the
desired gap, it may be necessary to use
spacers and to increase the size and
number of tack welds. Spacers are
usually short lengths of suitable diameter
clean stainless steel wire. Any cracked or
defective tack welds should be ground
out. Both ends of the tacks on open root
welds should be tapered to aid in fusing
into the root weld.
The need to maintain a proper gap
during root pass welding is two-fold.
First, a consistent and uniform gap aids
the welder in producing the optimum ID
root contour. When the joint closes up,
there is a tendency for concave roots
rather than the desired slightly convex
contour. The other reason for a uniform
root gap is the need to maintain the
optimum root pass chemical composition.
For many corrosion services, the filler
metal addition is essential to provide a
weld with corrosion resistance
comparable to the base metal. As the
joint closes, it is usually impossible to
melt a proper amount of filler metal into
the weld root. For example, the 6%
molybdenum stainless steels require
proper root gap and adequate filler metal
addition for high integrity root welds.
Purging during pipe root welding The
pipe interior must be purged with an inert
gas prior to the GTAW root pass. Failure to
use a purge can result in a heavily oxidized
ID root surface with substantially lower
corrosion resistance. Purging is usually with
pure argon, but nitrogen is sometimes used
because of lower cost. With duplex stainless
steels, nitrogen backing gas compensates
for nitrogen lost in the weld metal and restores weld pitting resistance. In Europe, a
nitrogen-10% hydrogen mixture is widely
used for purging austenitic steel pipe, but
would not generally be acceptable for duplex
steels.
Purging is a two-step operation, the first
being done prior to welding to displace air
inside the pipe. To save time and purging
gas, baffles on either side of the weld joint
are often used to reduce the purge area.

11

For the welder

GMAW processes. The areas covered in
earlier sections of this publication such
as base metal properties, joint designs
and preparation for welding are common
to all welding procedures and are not
repeated.

Shielded metal arc welding

SMAW is a versatile process, widely
used for welding stainless steel when
the shapes or quantity do not justify
automatic welding. The electrode is a
solid wire covered by an extruded flux
coating, although some manufacturers
use a cored wire in lieu of the solid core
wire. SMAW is frequently referred to as
covered electrode or stick welding. The
arc zone in the SMAW process is shown
in Figure 5.

Figure 4 Typical pipe purging fixtures.

Open root weld joints should be taped
and dead air spaces vented prior to
purging. The internal oxygen content
should be reduced to below 1 % prior to
welding. Typical purging fixtures are
shown in Figure 4.
Before the start of welding, the purge
flow rate should be reduced to the point
where there is only a slight positive
pressure. Tape covering weld joints
should be removed just in advance of
the area to be welded. After the root
pass, the internal purge should be
maintained during the next two filler
passes in order to minimize heat tint
(oxidation) on the inside weld surface.
This is especially important when it is
impractical to pickle after welding.
For those needing more information
on GTAW root pass pipe welding, there
are a number of technical articles and
specifications available. Two excellent
sources are the American Welding
Society publications listed in the General References.

Figure 5 Shielded metal arc welding.

The welding is performed manually
with the welder maintaining control over
the arc length and directing the arc into
the weld joint. The electrode coating
has these functions:
- the outer flux does not burn off as
fast as the electrode core which, in
turn, helps control the arc action and
ability to weld out-of-position;
- the flux is used to provide alloy
addition to the weld metal. The core
wire is not always the same
composition as the deposited weld
metal and therefore it is poor
practice to remove the flux and use
the core wire for filler with another
process such as GTAW;

Welding processes
This section provides information to
assist in formulating stainless steel
welding procedures for the shielded
metal arc welding, SMAW, GTAW and
12

For the welder

- the gaseous envelope from flux
decomposition excludes oxygen and
nitrogen from the molten weld metal;
- the molten slag formed on top of the
weld protects the weld metal from
contamination by the atmosphere
and helps to shape the bead.
Electrode types The electrodes are
selected first on the basis of weld metal
composition and then according to the
type of coating. Normally, they are of
matching or higher alloy composition to
the base metal. In some cases, it is an
engineering decision to use a special
composition electrode. The electrode
coating type is usually left to the individual fabricators. Electrodes for stainless steel base metals are shown in
Table IV.
The flux formula is usually each
manufacturer's jealously guarded
proprietary information. The flux coating
influences how the electrode operates in
various positions, shape and uniformity
of weld bead and that hard-to-define
operator appeal. There are two basic
classifications, namely -15 (lime) and 16 (basic-titania). For example, an
electrode may be either type 308-15 or
308-16. Electrode manufacturers often
establish their own suffix to designate
special electrodes but AWS A 5.4 - 81
recognizes only -15 and -16.
Lime coated electrodes (-15) are also
known as lime-fluorspar or basic type.
They are used on direct current, electrode positive, DCEP, (reverse polarity)
current, but some brands operate on
alternating current, AC. Lime coated
electrodes give the cleanest weld metal,
lowest in nitrogen, oxygen and inclusions. The weld metal tends to be
tougher, more ductile, more crack resistant and have the best corrosion resistance. The electrodes have good penetration and all-position weldability,
which is desirable for field work.
AC-DC coated electrodes (-16)
generally have a mixture of lime and
titania and are often used with alternat-

ing current. They are more popular than

the lime type because of better operating characteristics. The arc is stable
and smooth with a fine metal transfer.
The weld bead is uniform with a flat to
slightly concave contour. Slag is easily
removed without a secondary film
remaining on the weld bead.
Other guides in SMAW Factors
which contribute to high quality stainless
steel welds include proper handling and
storage of electrodes, correct welding
current along with good arc starting and
stopping techniques.

Electrode handling and storage

Stainless steel electrodes are normally
furnished in packages suitable for long
storage. After the package is opened,
the electrodes should be stored in
heated cabinets at the temperature
recommended by the manufacturer. If
the electrodes have been overexposed
to moisture, they should be reconditioned by a higher temperature bake
using the manufacturer's suggested
time and temperature. It is preferable to
obtain the manufacturer's specific
recommendations, since the temperature often varies with the particular
coating, but lacking this information,
commonly used temperatures are:
- storage of opened electrodes
225F (110C);
- recondition bake 500F (260C).
Moisture in the coating is a concern
because the hydrogen gas generated
can cause weld porosity. The pores may
be in the weld metal or may reach the
surface just as the metal solidifies,
forming visible surface pores. The
porosity can occur in butt welds when
the moisture content of the coating is
high, but more often occurs in fillet
welds. Excessive moisture in duplex
covered electrodes has the added risk of
causing hydrogen embrittlement in

13

Table IV
Suggested filler metals for welding stainless steels
Bare welding
electrodes and
rods AWS
or common
name
AWS A 5.9
(UNS)

Covered welding
electrode
AWS or common
name
AWS A 5.4
(UNS)

Bare welding
electrodes and
rods AWS
or common
name
AWS A 5.9
(UNS)

(3)

(3)

E 320LR
(W88022)

ER 320LR
(N08022)

AWSA5.9
(UNS)

Base
metal
AISI
(UNS)

Covered welding
electrode
AWS or common
name
AWS A 5.4
(UNS)

304
(S30400)

E 308
(W30810)

ER 308
(S30880)

20 MO-6
(N08026)

304L
(S30403)

E 308L
(W30813)

ER 308L
(S30883)

20Cb-3
(N08020)

309
(S30900)

E 309
(W30910)

ER 309
(S30980)

Castings

310
(S31000)

E310
(W31010)

ER 310
(S31080)

ACI type
(UNS)

AWSA5.4
(UNS)

316
(S31600)

E 316
(W31610)

ER 316
(S31680)

CF-8
(J92600)

E 308
(W30810)

ER 308
(S30880)

316L
(S31603)

E 316L
(W31613)

ER 316L
(S31683)

CF-3
(J92500)

E 308L
(W30813)

ER 308L
(S30883)

317
(S31700)

E 317
(W31710)

ER 317
(S31780)

CF-8M
(J92900)

E 316
(W31610)

ER 316
(S31680)

317L
(S31703)

E 317L
(W31713)

ER 317L
(S31783)

CF-3M
(J92800)

E 316L
(W31613)

ER 316L
(S31683)

317 LM
(S31725)

(3)

(3)

CN-7M
(J95150)

E 320 LR
(W88022)

ER 320 LR
(N08022)

321
(S32100)

E 347
(W34710)

ER 321
(S52180)

CK-3MCu
(S32154)

(3)

(3)

347
(S34700)

E 347
(W34710)

ER 347
(S34780)

CA-6NM
(J91540)

E 410 NiMo
(W41016)

ER 410 NiMo
(S41086)

Alloy 904L
(N08904)

(3)

(3)

Notes:

(1)

(1)

(1)

(1)

(2)

1925 hMo
(N08926)
(2)

25-6 Mo
(N08926)

(2)

(2)

(1)

(1)

(1)

(3)
(2)
(3)

(3)

AL-6XN
(N08367)

(2)

(1)

Alloy 254 SMO (3)

(S31254)
(2)

(1)

Base
metal
AISI
(UNS)

(3)

(3)

(3)

(3)

(3)

(1)

(1)

(1)

The L or low carbon grade or a stabilized grade is always used

for welded fabrication except in a few instances where the slightly
higher strength of the regular grades is more important than best
corrosion resistance.
Trade name.
A filler metal with 9% or more molybdenum such as the two listed
below is normally used to weld these stainless steels

Covered electrode
AWS A5.11
(UNS)
E NiCrMo-3
(W86112)
E NiCrMo-4
(W80276)

14

(1)

Bare welding
electrodes and rods
AWS 5.14
(UNS)
ER NiCrMo-3
(N06625)
Er NiCrMo-4
(N10276)

For the welder

Some of the advantages of this
process for welding stainless steels
include:
- no slag to remove which minimizes
post weld cleanup;
- an all position welding process which
is particularly useful in pipe welding;
- no weld spatter to clean;
- essentially no alloy loss during
welding.
GTAW equipment Direct current,
electrode negative, DCEN, (straight
polarity) current is standard. One option
is the pulsed-current where there is a
pulsating high rate of current rise and
decay. This current mode is well suited to
welding thin material and for joints which
have poor fit-up. Pulsed-current is also
useful in making the root pass of pipe
joints. A high-frequency starting feature is
commonly a part of the power source.
This allows an arc to be initiated without
a scratch start that may result in contamination of the tungsten electrode. Some
power sources are provided with a
feature that allows the electrode to be
positioned on the work but power does
not flow until the torch is lifted. An advantage over high frequency starting is that it
eliminates possible interference to nearby
components such as computers.
In addition to current controls at the
power source, it is often useful to have a
foot pedal current control. This control
allows the welder to increase or decrease
current during welding to adjust to
conditions such as poor fit-up. A further
advantage is at arc stops where slowly
reducing the current and in turn the weld
pool, effectively eliminates crater cracks.
Torches are either air or water cooled.
The air-cooled variety is limited to lower
currents than the water-cooled. The 2%
thoriated tungsten electrodes are most
commonly used because of their excellent emissive qualities, although other
tungsten electrode types are acceptable.
Opinions differ regarding electrode size
for various amperages. Some favour
using a different diameter for a number of
16

specific current ranges while others use

a size such as .09 in. (2.4 mm) for a
much wider current range. Also the
electrode end preparation preferences
vary but one commonly used is a 20 to
25 taper with the tip blunted to a 0.010
in. (0.25 mm) diameter.
Nozzle or gas cups come in a wide
variety of shapes and sizes and it is
often best to match the nozzle to the
weld joint or application. Larger cup
diameters provide better shielding gas
protection to the weld while smaller
nozzles help maintain a more stable arc
and allow better visibility. An alternate is
the gas lens which creates a laminar
flow by special screens inside the
nozzle. The flow of inert gas is projected
a considerable distance beyond the end
of the nozzle, giving both better gas
protection and good visibility.
With any welding process using inert
gas, it is important that all gas lines and
connections be checked to ensure
freedom from leaks in the system. If a
leak is present, for example in a gas
line, air will aspirate into the inert gas
stream rather than the internal gas
exiting as is sometimes believed.
Consumables Pure argon, helium
or mixtures of the two are used for
shielding gas in welding stainless steels.
The oxygen bearing argon mixtures used
in GMAW should not be used in GTAW
because of rapid deterioration of the
tungsten electrode. Nitrogen additions
are not recommended for the same
reason. In manual welding and joining
thicknesses below.06 in. (1.6 mm),
argon is the preferred shielding gas. It
provides good penetration at lower flow
rates than helium and less chance of
melt-through. Helium produces a higher
heat input and deeper penetrating arc
which may be an advantage in some
automatic welding applications. Argonhelium mixtures may improve the bead
contour and wetability.
The correct filler metals for GTAW
stainless steels are shown in Table IV.

For the welder

process are shown in Figure 7.

Gas metal arc welding

Arc transfer modes The type of

metal transfer in GMAW has a profound
influence on the process characteristics
to the extent that it is often misleading
to make general statements about
GMAW without indicating the arc
transfer mode. The three modes most
used in welding stainless steels are
spray, short circuiting and pulsed arc.
Table V compares some parameter and
usability differences in the three.

In the GMAW process (often referred

to as MIG when an inert shielded gas is
used or MAG when an active gas is
used), an arc is established between a
consumable, bare wire electrode and
the work piece. The arc and deposited
weld metal are protected from the
atmosphere by a gas shield, comprised
mainly of the inert gases, argon and/or
helium. Small amounts of active gases
such as carbon dioxide, oxygen and
hydrogen are optional for better wetting
and arc action. Some advantages of
GMAW over GTAW and SMAW include:
- faster welding speeds;
- no slag to remove which minimizes
post weld clean-up;
- ease of automation; and,
- good transfer of elements across the
arc.

GMAW equipment The same

power sources, wire feed mechanisms
and torches used for welding ordinary
steels are used for stainless steels.
Plastic liners in the wire feed conduit
have been found helpful in reducing
drag with stainless wire. The GMAW
process has more welding parameters
to control than GTAW and SMAW such
as amperage, voltage, current slope,

The basic components of the GMAW

Table V
Comparison of GMAW arc modes for stainless steels
Spray
arc welding
0.125 in. (3 mm) min.
0.25 in. (6 mm)
and thicker normal

Short circuiting
type transfer

Pulsed arc
welding

0.06 in. (1.6 mm)

and up

0.06 in. (1.6 mm)

and up

Welding positions

Flat & horizontal

all

all

Relative deposition
rate

highest

lowest

intermediate

Typical wire
diameter

0.06 in.
(1.16 mm)

0.030 or 0.035 in.

(0.8 or 0.9 mm)

0.035 or 0.045 in.

(0.9 or 1.2 mm)

Typical welding
current

250-300 amps

50-225 amps

up to 250 amps peak

Shielding gas(1)

Argon - 1 % O2
Argon - 2 % O2

90 % Helium
7.5 % Argon
2.5% CO2
or
90 % Argon
7.5 % Helium
2.5%CO2

90 % Helium
7.5 % Argon
2.5 % CO2
or
90 % Argon
7.5 % Helium
2.5%CO2
or
Argon -1 % O2

Typical thickness
welded

(1) Other gas mixtures are used, however, the shielding gas should contain at least 97.5 % inert gas,
i.e., argon, helium or a mixture of the two.

18

For the welder

wire feed, pulse rate and the arc
transfer mode. Consequently the
GMAW power sources are often more
complex and expensive. Some of the
newer power sources such as the
synergic pulsed arc have made operation simpler by providing only one
control dial for the operator, with other
parameters adjusted automatically. The
welding current used more than 95% of
the time is DCEP (reversed polarity).
This current gives deeper penetration
than DCEN (straight polarity) and a
stable arc. DCEN is limited to applications requiring shallow penetration
such as overlay welding.

advances in flux cored arc products

which produce quality welds at higher
efficiency than SMAW. Cored wires are
often easier to produce to special
compositions or ferrite ranges than it is
to melt large heats for solid wire.
Submerged arc welding, SAW, has
been used extensively for welding
thickness about 0.25 in. (6.4 mm) and
thicker and for overlay welding. Commercial fluxes are available for use with
standard filler metals used for GMAW.
Plasma arc, electroslag, electron
beam, laser and friction welding are
used more and more and the
resistance welding processes; spot,
seam, projection and flash welding are
readily adaptable to stainless steels.
Stainless steel may be joined to itself
or a number of other metals by brazing.
It is not usually used when the joint will
be exposed to severe corrosive environments but there are instances in
food and other process industries
where brazing provides adequate
properties.
Oxyfuel welding, OFW, is not recommended for stainless steels. The
chromium oxide formed on the surface
makes oxyacetylene welding difficult.
However, more important is the extreme care needed in welding to avoid
reducing the corrosion resistance of the
weld and weld area.

Consumables Some of the more

common shielding gases used in
GMAW are shown in Table V. Spray arc
shielding gas is usually argon with
either 1 % or 2% oxygen. Short
circuiting and pulsed arc welding use a
greater variety of shielding gases. A
popular mixture in North America is
90% helium, 7.5% argon and 2.5% CO2
but in Europe, helium is quite expensive
and 90% argon, 7.5% helium and 2.5%
CO2 is widely used. Whatever the
combination, the shielding gas should
contain at least 97.5% inert gases
(argon, helium or a mixture of the two).
Carbon dioxide should not exceed 2.5%
or the weld quality and corrosion
resistance may be reduced.
The preferred filler metals to be used
in GMAW stainless steels are shown in
Table IV. The most widely used diameters are 0.035 in. 0.045 in. and 0.062
in. (0.9 mm, 1.2 mm and 1.6 mm) but
others are available.

Post-fabrication cleaning
All too often, it is assumed the fabrication, be it a tank, pressure vessel, pipe
assembly etc., is ready for service after
the final weld is made and inspected.
Post-fabrication cleaning may be as
important as any of the fabrication steps
discussed above. The surface condition
of stainless steels is critical, both where
the product must not be contaminated,
e.g., pharmaceutical, food and nuclear
plants; and where the stainless must
resist an aggressive environment such
as in a chemical or other process
industry plant. Surface conditions that

Other welding processes

Stainless steels can be welded by
most of the commercial welding processes. These processes may offer
advantages not obtainable in SMAW,
GTAW and GMAW processes and
should not be overlooked for high
production or special fabrications. As an
example, there have been recent
19

For the welder

can reduce corrosion resistance may be
grouped into four categories; surface
contamination, embedded iron,
mechanical damage or welding related
defects. Figure 8 illustrates some of the
common conditions.

must be free of organic contaminants

for the acid to be effective in removing
free iron, surface oxides or similar
conditions. Because little can be done
during fabrication to reduce organic
contamination, the fabricator must do
this during the final cleanup.
Detection Visual inspection is
usually used for organic
contamination, while cloth or paper
can be used for oil or grease detection.

Removal Degreasing, using a

nonchlorinated solvent, is effective. The
water-break test is a simple way to judge
the effectiveness of degreasing. A thin
sheet of water (applied by a hose)
directed on a vessel wall will break
Figure 8 Typical fabrication defects or
around oil, grease or similar surface
surface conditions commonly encountered.
contamination. Degreasing should be
redone until the water stops breaking.
A chlorinated solvent is not recommended because residual chlorides may
Surface contaminants
remain in crevices and cause crevice
In aggressive environments, organic
corrosion or chloride stress corrosion
contaminants on stainless steel surcracking later when the unit is placed in
faces can foster crevice corrosion.
service.
Such contaminants include grease, oil,
crayon marks, paint, adhesive tape and
Embedded iron
other sticky deposits. Figure 9 shows
Sometimes, new stainless-steel tanks
crevice corrosion pits (in the area
or vessels rust shortly after delivery
marked 33) on a stainless steel vessel.
from a fabricator. This may be due to
The pits formed where crayon markings
iron embedded in the surface during
were not removed from the surface
fabrication. The iron particles corrode in
before the vessel was put in service.
moist air or when wetted, leaving telltale
Surfaces to be pickled or acid treated

Figure 10 A deep scratch made during

Figure 9 Crevice corrosion occurred
where crayon marks were made and not fabrication served as an initiation site for
corrosion in this vessel.
removed on a stainless steel vessel.
20

For the welder

rust streaks. In addition to being unsightly as they corrode, the larger
particles of embedded iron may initiate
crevice corrosion in the underlying
stainless steel. Figure 10 shows corrosion at several points along a scratch
where iron had been embedded. In
corrosive service, crevice corrosion
initiated by large embedded iron particles may lead to corrosion failure that
would not otherwise occur. In the
pharmaceutical, food, and other
processing industries in which stainless
is used primarily to prevent contamination of the product, embedded
iron cannot be tolerated.
Detecting embedded iron The
simplest test for embedded free iron is
to spray the surface with clean water
and drain the excess. After 24 hours,
the surface is inspected for rust streaks.
This is a minimum test that any fabricating shop can conduct. To ensure
against rust-streaked units, the water
test should be specified in procurement
documents.
A more sensitive indication of embedded iron is obtained by use of the
ferroxyl test for free iron. The test
solution is prepared by mixing the
following ingredients:
Ingredient
%
Distilled water
94
Nitric acid, 60-67%
3
Potassium ferrocyanide 3

Amount
volume or weight
3

1,000 cm
3
30 cm
30 g

The solution is best applied using a

one-quart spray applicator, the type that
applies bleach to laundry. Iron contamination is indicated by the appearance of
a blue colour after a few minutes. The
depth of colour roughly indicates the
degree of contamination. The solution
should be removed after a few minutes
with a water spray or a damp cloth.
The ferroxyl test is not only sensitive
but it can be performed in the field as
easily as in the shop. Personnel can be

trained to administer it in only a few

hours. This test is generally required
for stainless steel equipment used in,
for example, pharmaceutical, food and
nuclear plants, as well as for equipment used to process chemicals. An
excellent basic guide to these tests is
ASTM A380, Standard Recommended Practice for Cleaning and
Descaling Stainless Steel Parts.
Removing embedded iron
Pickling, which is carried out after
degreasing, is the most effective
method for removing embedded iron.
In pickling, the surface layer, less than
0.001 in. (0.025 mm), is removed by
corrosion, normally in a nitric/
hydrofluoric acid bath at 120F,
(50C). Pickling not only removes
embedded iron and other metallic
contamination, it leaves the surface
bright and clean, and in its most
resistant condition. Since pickling is
controlled corrosion, low-carbon or
stabilized grades of stainless are
preferred. The process may initiate
intergranular corrosion in the HAZ of
regular unstabilized grades. Because
pickling is aggressive, it will destroy a
polished or high-luster surface.
Using nitric acid alone will remove
superficial iron contamination but few,
if any, of the larger, deeply embedded
particles. Nitric acid treatment is
referred to as passivation. This can be
misleading, since the pickled surface
is also passivated when it contacts
air.
Small objects are best pickled by
immersion. Piping, field-erected tanks
and vessels too large to immerse can
be treated by circulating the pickling
solution through them. Typically,
chemical-cleaning contractors are
hired to do this.
When ferroxyl testing shows only
spotty patches of iron, these can be
removed by local application of nitric/
hydrofluoric acid paste. For large
tanks, filling to about 6 in. (150 mm)
21

For the welder

to pickle the bottom, and locally removing embedded iron on side-walls is often
a practical alternative to circulating
pickling solution throughout them.
When pickling is not practical, blasting
can be used, but not all abrasives yield
good results. Glass-bead blasting
produces good results but, before
blasting, a test should be made to
determine that it will remove the surface
contamination. Also, periodic tests
should be made to see how much reuse
of beads can be tolerated before they
begin to recontaminate the surface.
Walnut shells have also performed well
as an abrasive.
Abrasive blasting with steel shot or grit
is generally unsatisfactory because of
the possibility of embedding iron
particles. Also, grit blasting leaves a
rough profile that makes the stainless
steel susceptible to crevice corrosion,
whether or not the surface is free of iron.
Sand blasting should also be avoided if
possible, because even new sand is
seldom free of iron particles or other
contaminants.

GTAW is usually used because of greater

ease in making small repair welds. Filler
metal should always be added and wash
passes or cosmetic welds never allowed
because of the risk of weld cracking and
reduced corrosion resistance.

Safety and welding fumes

Safety rules for welding stainless steels

are essentially the same as for all metals
as they pertain to areas such as electrical equipment, gas equipment, eye and
face protection, fire protection, labeling
hazardous materials and similar items. A
good reference guide on welding safety
is American National Standard Institute /
Accredited Standards Committee, ANSI/
ASC, Z49.1-88, Safety in Welding and
Cutting, published by the American
Welding Society.
Proper ventilation to minimize the
welders exposure to fumes is important
in welding and cutting all metals, including stainless steels. In addition to good
ventilation, the welders and cutters
should try to avoid breathing the fume
plume directly, by positioning the work so
that their head is away from the plume.
Mechanical damage
The composition of welding fumes varies
When the surface has been damaged
with the welding filler metal and welding
and reconditioning is needed, the repair process. Arc processes also produce
is usually made by grinding or by
gaseous products such as ozone and
welding and grinding. Shallow defects
oxides of nitrogen. Concern has been
are first removed by grinding, preferably expressed in welding with stainless steel
with a clean fine grit abrasive disk, a
and high alloy steel consumables beflapper wheel or a pencil type grinder.
cause of the chromium and, to a lesser
The maximum grinding depth to remove extent, the nickel usually present in the
defects is often specified by the fabrica- welding fume. Good ventilation will
tion specification and may vary from
minimize the potential health risk. The
10% to 25% of the total thickness.
International Institute of Welding has
When weld repair is needed, it can be developed a series of Fume information
made by SMAW, GMAW or GTAW.
sheets for welders which offer internationally accepted suggested guidelines
for fume control.

22

Part II
For the materials engineer
This section is for the engineer who
needs further information about the
wrought and cast stainless steel
alloys, how their corrosion resistance
is affected by welding and typical heat
treating practices. Also included are
guides for material procurement and
good storage practices.

Stainless steel alloys

Steel is made corrosion resistant by the
addition of 11 % or more chromium. The
term stainless describes the non-rusting,
bright appearance of these alloys. The
earliest types of stainless steel were the
straight chromium grades with
chromium ranging from over 10% to
about 18%, but through the years a
number of different types of stainless
steel alloys have been developed and
categorized into five groups, namely:
- martensitic
(AISI* 400-series)
- ferritic
(AISI* 400-series)
- austenitic
(AISI* 300-series)
- precipitation hardening
- duplex
*American Iron and Steel Institute

The austenitic stainless steels are the

most widely used but the use of duplex
alloys is increasing, although they still
represent a small part of the stainless
steels used. This publication describes
these two alloy families and their use.
The other three groups, martensitic,
ferritic and precipitation hardening are
also identified as stainless steels but the
fabrication and welding is often quite
different from the austenitic and duplex
grades. When discussing welding and
fabrication techniques, the particular
stainless steel group must be identified,
otherwise serious mistakes could be
made. For example, using a procedure

developed for an austenitic stainless

steel to weld a martensitic stainless
steel could result in low quality welds.

Austenitic stainless steels

Austenitic stainless steels are nonmagnetic or only slightly magnetic in the
annealed state and can be hardened
only by cold working. They possess
excellent cryogenic (low temperature)
properties and good strength at high
temperatures. Corrosion resistance is
outstanding in a wide range of environments. They exhibit good weldability and
are easy to fabricate provided suitable
procedures are maintained.
The composition of common grades of
wrought stainless steels and corrosionresistant stainless steel castings is
shown in Tables Vl and Vll. It includes
alloys commercially available worldwide
and those most frequently used for
corrosion resistant applications. The
UNS numbers in Table Vl have either an
S or N prefix. Stainless steels are
defined by ASTM as having at least 50%
iron which UNS identifies with a S
number. Alloys with a N number are
classified as nickel alloys, but the
distinction is purely artificial. The
fabricability of the high alloy S grades
and the nickel alloys in Table VI is
essentially the same.

Effect of welding on corrosion

resistance
Austenitic stainless steels are usually
specified for their excellent corrosion
resistance. Welding can reduce base
metal corrosion resistance in aggressive
environments. In welding, heat is
generated that produces a temperature
gradient in the base metal, i.e. the HAZ.
23

For the materials engineer

Table VI
Wrought austenitic stainless steels chemical analysis, %,
of major elements
(Max. except as noted)
AISI Type
or common name
(UNS)

Cr

Ni

Mo

Other

304
(S30400)

0.08

18.0-20.0

8.0-10.5

0.10N

304 L
(S30403)

0.03

18.0-20.0

8.0-12.0

0.10N

309
(S30900)

0.20

22.0-24.0

12.0-15.0

310
(S31000)

0.25

24.0-26.0

19.0-22.0

316
(S31600)

0.08

16.0-18.0

10.0-14.0

2.0-3.0

0.10N

316L
(S31603)

0.03

16.0-18.0

10.0-14.0

2.0-3.0

0.10N

317
(S31700)

0.08

18.0-20.0

11.0-15.0

3.0-4.0

0.10N

317L
(S31703)

0.03

18.0-20.0

11.0-15.0

3.0-4.0

0.10N

317 LM
(S31725)

0.03

18.0-20.0

13.0-17.0

4.0-5.0

0.10N

321
(S32100)

0.08

17.0-19.0

9.0-12.0

5 x %C min,
0.70 max. Ti

347
(S34700)

0.08

17-0-19.0

9.0-13.0

Alloy 904L
(N08904)

0.02

19.0-23.0

23.0-28.0

4.0-5.0

10 x %C min,
1.10 max.
(Nb + Ta)
1.0-2.0 Cu

Alloy 254 SM0*

(S31254)

0.02

19.5-20.5

17.5-18.5

6.0-6.5

0.18-0.22N
0.50-1.00 Cu

AL-6XN*
(N08367)

0.03

20.0-22.0

23.5-25.5

6.0-7.0

0.18-0.25N
0.75 Cu

1925 h Mo*
(N08926)

0.02

20.0-21.0

24.5-25.5

6.0-6.8

0.18-0.20N
0.8-1.0 Cu

20 Mo-6*
(N08026)

0.03

22.0-26.0

33.0-37.0

5.0-6.7

2.0-4.0 Cu

20Cb-3*
(N08020)

0.07

19.0-21.0

32.0-38.0

2.0-3.0

25-6 MO*
(N08926)

0.02

19.0-21.0

24.0-26.0

6.0-7.0

* 254 SMO is a trademark of Avesta AB

AL-6XN is a trademark of Allegheny Lundlum Steel Corporation
1925 hMo is a trademark of VDM Nickel Technologie A.G.
20 Mo-6 and 20 Cb-3 are trademarks of Carpenter Technology Corp.
25-6 MO is a trademark of Inco Alloys International, Inc.

24

3.0-4.0 Cu
8 x C min, 1.00 %max.
0.15-025N
0.5-1.5 Cu

For the materials engineer

Table VII
Corrosion resistant stainless steel castings chemical analysis, %,
of major elements
(Max. except as noted)
ACI
Type
(UNS)
CF-8
(J92600)

Similar
wrought
type
304

C
0.08

Cr
18.0-21.0

Ni
8.0-11.0

Mo
-

Others
-

CF-3
(J92500)

304L

0.03

17.0-21.0

8.0-11.0

1900-2050
(1035-1120)

CF-8M
(J92900)

316

0.08

18.0-21.0

9.0-12.0 2.0-3.0

1900-2050
(1035-1120)

CF-3M
(J92800)

316L

0.03

17.0-21.0

9.0-13.0 2.0-3.0

1950-2050
(1065-1120)

CN-7M
(N08007)

20Cb-3 (1)

0.07

19.0-22.0

27.5-30.5 2.0-3.0

3.0-4.0Cu

CK-3M Cu
(J93254)

Alloy254SMO(2) 0.02

19.5-20.5

17.5-18.5 6.0-6.5

0.06

11.5-14.0

3.5-4.5

FERRALIUM (3) 0.04

255

24.0-27.0

4.5-6.5 2.0-4.0

0.10-0.25N Duplex1.5-2.5 Cu austenite &

ferrite

2205

0.03

21.0-23.5

4.5-6.5 2.5-3.5

0.10-0.30N

Zeron 100 (4)

0.03

24.0-26.0

6.0-8.5 3.0-4.0

0.2-0.3N

CA-6NM
(J91540)

CD-7MCuN

CD-3MN
ASTM-A890, Gr4A
(J92205)
Zeron 100
(4)
(J93380)

0.4-1.0

Most
common
structure
Ferrite in
austenite

Anneal at
F
(C)
1900-2050
(1035-1120)

Austenite

2050 (1120)
Min.

0.18-0.22N Austenite
0.50-1.00 Cu

2100-2200
(1150-1205)

Martensite

1900-1950
(1035-1065)
followed by
double
temper
1925 (1050)
Min.

Duplexaustenite &
ferrite

2050 (1120)
Min.

Duplexaustenite &
ferrite

2050 (1120)
Min.

(1) 20Cb-3 is a tradename of Carpenter Technology Corporation

(2) 254SMO is a trademark of Avesta AB

(3)FERRALIUM is a trademark of Langley Alloys, Ltd

(4) Zeron 100 is a trademark of Weir Material Services, Ltd.

Welding may also induce residual

stresses in the weld area which in
certain environments can lead to
SCC. Heat treatments to reduce
residual stresses are discussed in the
section Heat Treatment of Austenitic
Stainless Steels.
One of the early corrosion problems
related to welding was intergranular

attack, IGA, in the weld HAZ. In the

temperature range of about 800F to
1650F (425C to 900C), carbon
combines with chromium to form chromium carbides at the grain boundaries.
The area adjacent to the carbides is
depleted in chromium. When the carbide network is continuous, the low
chromium envelope around grains may

25

For the materials engineer

Figure 11 Effect of carbon control on carbide precipitation in Type 304.

chromium carbides have formed at the
grain boundaries, when the time at
temperature for any particular carbon
content Type 304 is to the right of the %
carbon curve. It can be seen that the
temperature where sensitization occurs
most rapidly varies from 1300F
(700C), with an alloy of 0.062% carbon, to 1100F (600C), for a 0.03%
carbon alloy. From Figure 11, an alloy
with 0.062% carbon could become
sensitized in as little time as 2 to 3
minutes at 1300F (700C ). On the
other hand, 304L with 0.030% carbon
could be held at 1100F (595C) for 8
hours before being sensitized. For this
reason the low carbon L grades are
most commonly used for corrosion
resistant equipment where IGA is a
possibility. With the L grade, the weld
HAZ is not at temperature long enough
to become sensitized.
Grain boundary chromium carbides
can be prevented from forming when
titanium or niobium-tantalum are
present in the alloy. (Niobium,Nb, is
also known by the name columbium,
Cb, in some references). These ele-

be selectively attacked, resulting in

intergranular corrosion. In the worst
case, the depleted chromium layer is
corroded away and complete grains are
separated from the base metal and can
even fall out. Alloys are said to be
sensitized when welding or heat treatment results in chromium depleted
areas that will be attacked in these
corrosive environments. Sensitized
alloys may still provide good service in
many of the milder environments in
which stainless steels are used. Today,
with the trend of mills to furnish lower
carbon products, IGA of austenitic
stainless steels occurs less often.
The degree of sensitization, i.e., the
amount of grain boundary chromium
carbides formed, is influenced by the
amount of carbon, exposure temperature and time at this temperature. Figure
11 illustrates the time-temperaturesensitization curves for Type 304
stainless steel. Curves of other
austenitic stainlesses would be similar
but the actual values would be somewhat different. To explain Figure 11, the
alloy is sensitized, that is a network of
26

For the materials engineer

cooling, the higher the ferrite content.
Unfortunately, ferrite is not obtainable
in all nickel stainless steel alloys. For
example, it is not possible to adjust the
composition to obtain ferrite in Type 310
(UNS S31000). In spite of being fully
austenitic and prone to fissures, the
alloy has been used over 50 years with
excellent service. In the absence of
weld metal ferrite, it is more important
for the filler metal manufacturer to
control minor elements such as silicon,
phosphorus and sulphur to as low a
level as possible to prevent cracking.
When a filler metal is required with a
specific ferrite level, the purchaser or
user should specify the level to the
supplier. Stainless steel filler metal
specifications, ANSI/AWS A5.4 for
electrodes and ANSI/AWS A5.9 for bare
wire do not specify ferrite levels for any
of the alloy classes.

ments have a greater affinity for carbon

than chromium and form evenly distributed carbides away from the grain
boundaries, where there is no adverse
affect on the corrosion resistance. Type
321 (UNS S32100) contains titanium
and 347 (UNS S34700) contains niobium-tantalum. Both are stabilized
versions of type 304. The stabilized
grades are usually preferred where
there will be long time service in the
sensitizing temperature range of 800F
to 1650F (425C to 900C).
A third method of preventing IGA in
the weld HAZ in alloys containing over
0.03% carbon is to redissolve the
chromium carbides by a solution anneal
at 1900F to 2150F (1040C to
1175C), followed by rapid cooling.
The solution anneal is a good method to
restore full corrosion resistance when
the shape, size and geometry of the
weldment allows the heat treatment.
Solution annealing must be closely
controlled in both heating and cooling to
minimize distortion within acceptable
limits.

Role of weld metal ferrite

Microfissures or cracks have been
known to occur in austenitic stainless
steel welds. They can appear in the weld
metal during or immediately after
welding, or they may occur in the HAZ of
previously deposited weld metal. The
microstructure of the weld metal strongly
influences susceptibility to
microfissuring. A fully austenitic weld is
more prone to microfissuring than a weld
with some ferrite.
Ferrite levels of 5% to 10% or more in
welds or castings can be quite beneficial
in reducing hot cracking and
microfissuring. For example, a Type 308
(UNS W30840) weld with zero to 2%
ferrite might be quite crack sensitive
while another 308 weld with 5% to 8%
ferrite would have good crack resistance.
The amount of ferrite in a 300-series
weld is controlled by the composition
and the weld cooling rate, the faster the
27

Measuring weld metal ferrite

While there is wide agreement on the
beneficial affect of ferrite in the weld,
it is not always easy to measure the
amount accurately in a given weld
deposit. One of the three following
methods can be used.
1. Magnetic instruments can measure
ferrite on a relative scale and this is the
method most used by filler metal producers. Calibration of the instruments is
very critical and AWS has developed a
special calibration procedure. AWS also
details how the weld pad is to be made
and prepared for testing, since this can
influence the measurement. Ferrite
determination using sophisticated
laboratory magnetic instruments is often
not practical for the average user.
Portable magnetic instruments are
commercially available that, even
though they may be less accurate, are
easier for the fabricator to use.
2. Using the weld chemical composition, ferrite content can be estimated
from a constitution diagram for stainless

For the materials engineer

Figure 12 Revised constitution diagram for stainless steel weld metal.

(from Metals Handbook, Volume 6, Ninth Edition)
steel weld metal, Figure 12. Earlier,
ferrite diagrams represented ferrite in
units of volume-%. The most recent
Welding Research Council, WRC,
diagram determines ferrite number, FN,
by the magnetic response. The FN and
volume-% are the same up to 6% but
differ at higher levels. Ferrite determination using the diagram is easy and quite
accurate, provided a reliable chemical
analysis has been made.

temperatures 900F to 1700F (480C

to 925C) to avoid a loss of room
temperature ductility as a result of a
high temperature sigma phase. Sigma
forms more readily from ferrite than
from austenite and is discussed in the
duplex stainless steel section.

Duplex stainless steels

Duplex stainless steels are an alloy
family that have two phases ferrite
and austenite with ferrite typically

3. The ferrite content can be estimated

by metallographic examination. It is
most accurate when ferrite is in the
range of 4% to 10% and should be
performed by an experienced technician. One advantage to this method is
that it can be used on small specimens
removed from weldments or where the
two other methods are not practical.
There are services where ferrite in
the weld structure is not a benefit. At
cryrogenic temperatures,viz., -320F
(-195C), toughness and impact
strength are reduced by ferrite and it is
common practice to specify welds with
no more than 2 FN and preferably 0 FN.
It is also desirable to have low ferrite
when the welds are exposed to service

Figure 13 Typical microstructure of cold

rolled, quench annealed Alloy 2205
seamless tube. Dark phase ferrite,
light phase austenite.
(from Sandvik AB)
28

For the materials engineer

is improved pitting and crevice corrosion
resistance. With proper welding procedures, as-welded second-generation
duplex stainless steels can have nearly
the same level of corrosion resistance as
mill annealed material. Nitrogen is also
beneficial in the manufacture of secondgeneration alloy plates, where the
ductile-brittle transition is depressed well
below room temperature, making heavy
section weldments practical. However,
duplex alloys are generally not used
below about -50F (-45C) whereas
some fully austenitic alloys may be used
to -456F (-270C).
Alloy 2205 (UNS S31803) is the most
widely used duplex alloy and is available
from a number of producers. Comparing
the duplex composition to a fully
austenitic stainless steel such as Type
316, 2205 is higher in chromium, lower
in nickel and contains nitrogen. The
nitrogen addition is very critical in duplex
alloys as will be discussed shortly.

between 40% and 60%.The ferrite/

austenite ratio is accomplished in
wrought alloys by composition adjustment along with controlled hot working
and annealing practices at the mill. The
alloys could properly be called ferriticaustenitic stainless steels but the term
duplex is more widely used. A typical
duplex stainless steel microstructure is
shown in Figure 13. The matrix which
appears as the darker background is
ferrite and the elongated, island-like
lighter phase is austenite.
Duplex alloys date to the 1930s and
the early alloys are now identified as
first-generation. Unfortunately, the early
alloys had a problem of significant loss
of corrosion resistance in the as-welded
condition and it has taken some time for
the new second-generation alloys to
overcome this reputation. All the alloys
shown in Table Vlll are second-generation alloys and typically contain 0.15% to
0.30% nitrogen. One benefit of nitrogen

Table VIII
Duplex stainless steels chemical analysis, %,
of major elements
(Max. except as noted)
Common name
(UNS)
(1)
7-Mo PLUS
(S32950)

C
0.03

Cr
26.0-29.0

Ni
3.5-5.2

Mo
1.0-2.5

Others
0.10-0.35N

Alloy 2205
(S31803)

0.03

21.0-23.0

4.5-6.5

2.5-3.5

0.08-0.20N

0.03

24.0-27.0

4.5-6.5

2.0-4.0

0.10-0.25N
1.5-2.5Cu

0.03

24.0-26.0

6.0-8.0

3.0-5.0

0.24-0.32N

0.03

24.0-26.0

6.0-8.0

3.0-4.0

0.5-1.0 Cu
0.5-1.0 W
0.2-0.3 N

(2)

FERRALIUM 255
(S32550)
(3)

SAF 2507
(S32750)

(4)

Zeron 100
(S32760)
(1)
(2)
(3)
(4)

7-Mo PLUS is a trademark of Carpenter Technology Corporation

FERRALIUM is a trademark of Langley Alloys, Ltd.
SAF 2507 is a trademark of Sandvik AB
Zeron 100 is a trademark of Weir Material Services, Ltd.

29

For the materials engineer

producing or fabricating the alloys, a
high temperature solution anneal at
1900F (1040C) or higher, depending
on the alloy, followed by rapid cooling is
employed to give optimum mechanical
properties and corrosion resistance. In
exposure to the temperature range of
600F to 1750F (315C to 950C) the
duplex alloys act differently than the
austenitics but once the differences are
recognized, no problems should arise.
An intermetallic phase called sigma
can form when duplex alloys are held in
the 1200F to 1750F (650C to 950C)
temperature range. Sigma causes room
temperature embrittlement and, when
present in appreciable amounts, corrosion resistance is lowered. However,
attention to minimum time in the sigma
forming range during annealing and
welding, improved processing control at
the steel mill and the beneficial effect of
nitrogen can essentially eliminate any
sigma problem. In normal secondgeneration duplex welding procedures,
the weld or HAZ is not at temperature
long enough for sigma to be a factor.
Another high temperature occurrence is
a phenomenon called 885F (475C)
embrittlement. It can occur when a
duplex alloy (or any iron-chromium alloy
containing 13% to 90% Cr) is held within
or slowly cooled through the
temperature range of 600F to 1000F
(315C to 540C). With the secondgeneration duplex alloy and using
standard annealing and welding practices, the weld or HAZ is not at temperature long enough for this embrittlement
to occur. It is mentioned here as a
precaution should there be need to
deviate from standard procedures.

Characteristics of duplex
stainless steels
The duplex alloys offer two important
advantages over austenitic alloys such
as 304L and 316L, namely greater
resistance to chloride stress corrosion
cracking, CSCC, and higher mechanical
properties. The yield strength of duplex
alloys is typically two to three times
higher and the tensile strength 25%
higher while still maintaining good
ductility at normal operating temperatures.
The susceptibility of austenitic stainless steels to CSCC at temperatures
above about 140F (60C) is a well
known concern. The ferritic stainless
steels are highly resistant but are more
difficult to fabricate and weld. The
duplex alloys have intermediate resistance to CSCC which, in many environments, represents a substantial improvement over the austenitics. The
duplex alloys also offer:
- general and pitting corrosion resistance equal to or better than type
316L stainless steel in many environments;
- resistance to intergranular corrosion
due to the low carbon content;
- good resistance to erosion and
abrasion; and,
- a thermal expansion coefficient close
to that of carbon steel which can
result in lower stresses in weldments
involving duplex stainless and
carbon steel.
There are metallurgical differences
compared to the austenitic alloys that
when known and recognized are easily
handled. The differences occur as a
result of high temperature exposure.

Effect of welding on duplex

stainless steels
The weldability of second-generation
duplex alloys has been greatly improved
through controlled nitrogen additions
and the development of nickel-enriched
filler metals. Using a few welding
procedure controls, sound welds with

High temperature exposure

Duplex stainless steels are normally
used in the temperature range of about
-50F to 500F (-45C to 260C). In
30

corrosion resistance comparable to the

base metal are obtained. The importance of controls on heat input, interpass
temperature, preheat and nickel-enriched filler metal are as follows:
Nickel-enriched filler metal Duplex
stainless steel welds made with matching
composition filler metal or autogenously
welded (no filler metal) may exhibit 80%
or more ferrite in the fusion zone in the
as-welded condition. A weld with such a
high ferrite level has poor toughness and
ductility and often will not pass a bend
test. The higher ferrite content of such
welds also markedly reduces corrosion
resistance in many aggressive environments. An anneal at 1900F to 2100F
(1040C to 1150C) restores the desired
ferrite/austenite balance but the treatment is not practical for many
fabrications and is expensive. Increasing
the nickel content of the filler metal
allows more austenite to form so that
welds in the as-welded condition have
typically 30% to 60% ferrite. Welds made
with nickel-enriched filler metals have
good as-welded ductility, are able to
pass bend tests, and have corrosion
resistance comparable to the base metal.
It is desirable that all weld passes be
made with substantial filler metal addition
to provide a nickel enhanced weld metal
compostition. A large amount of base
metal dilution can result in welds having
a high ferrite content with lower ductility
and toughness. An example of where
this can occur is the root pass of a pipe
weld with high base metal dilution.
Special care should be taken to add
sufficient nickel-enriched filler metal.
Joints with a feather edge and tight fitups favor high dilution and are best
avoided. Joints with an open root
spacing and a land are preferred since
they require the addition of filler metal.
Nickel-enriched filler metal products
for the duplex alloys are available as
covered electrodes, bare filler metal and
flux cored wire as shown in Table IX.
Duplex filler metals are not covered by

For the materials engineer

current AWS stainless steel filler metal
specifications but will be included in
future editions.
Heat input control There is not
complete agreement on the part of
producers and welding investigators as
to the proper limits on heat input. The
argument for high heat input (see
formula) is that it allows more time for
ferrite to transform to austenite, particularly in the heat affected zone. The
concern for high heat input is that it
could allow embrittling phases, such as
sigma and 885F (475C) embrittlement
to develop in the ferrite: With the
second-generation duplex stainless
steels, longer time at temperature is
needed for these phases to develop, so
there should be no significant
embrittlement. A generally accepted
heat input range in kilo joules (kJ) is 15
to 65 kJ/in. (0.6 to 2.6 kJ/mm) although
levels as high as 152 kJ/in. (6.0 kJ/mm)
are claimed to have been successfully
used. When a welding process with
less than 15 kJ/in. (0.6 kJ/mm) heat
input must be used, preheating to
200F-400F (95C - 205C) is helpful in
reducing the cooling rate and increasing
austenite in the weld. Where there is a
question on heat input for a particular
duplex alloy, it is a good practice to
contact the material supplier for specific
recommendations.
Heat input in kJ/in. is calculated:
Voltage x Amperage x 60
Travel speed (inch/minute) x 1000
Interpass temperature control An
early concern was that a high interpass
temperature could result in 885F
(475C) embrittlement and a limit of
300F (150C) maximum interpass
temperature was suggested. This limit
is conservative and in some instances a
maximum limit of 450F (230C) could
be acceptable. However, in the interest
of consistency, fabricators often specify
31

For the materials engineer

Table IX
Duplex stainless steel filler metals
typical composition
Filler metal
common name
(UNS)
Covered electrodes
(1)
2209-16
(W39209) tentative

For welding
base metal
C

Bare filler wire

(2)
22.8.3L

7-Mo PLUS Enriched Ni

(4)

FERRALIUM 255
(S39553) Tentative
Zeron 100 filler wire

(5)

(3)

23

9.7

3.0

0.10N

0.03

22

9.5

0.15N

7-MO PLUS
(S32950)

0.03

26.5

9.5

1.5

0.20N

0.03

25

7.5

3.1

0.20N
2.0 Cu

0.01

22.5

0.10N

7-Mo Plus
(S32950)

0.02

26.5

8.5

1.5

0.20N

FERRALIUM 255
(S32550)

(4)

FERRALIUM 255
(W39553) tentative

0.03

3 RE60 (S31500)
2205(S31803)
2304(S32304)

(3)

Others

FERRALIUM 255
(S32550)

7-Mo PLUS Enriched Ni

Mo

3RE60 (S31500)
2205 (S31803)
2304(S32304)

(2)

Ni

2205
(S31803)

22.9.3.L-16
(2)
22.9.3.L-15
(2)
22.9.3.LR

Cr

0.03

25

5.8

3.0

0.17N

Zeron 100
(S32760)

0.03

25

10*

3.5

0.25N
0.7 Cu
0.7 W

2205
(S31803)

0.02

22.0

8.5

3.3

0.14N

FERRALIUM 255
(S32550)

0.02

25

10

3.2

0.14N
2.0 Cu

Flux Cored Wire

In-Flux 2209-0
(W31831)
In-Flux 259-0

(1)

(2)

(3)

(1)

(1)

2209-16, In-Flux 2205-0 and In-Flux 259-0 are

trademarks of Teleldyne McKay
22.9.3L-16, 22.9.3L-15, 22.9.3.LR and 22.8.3L are
trademarks of Sandvik AB
7-Mo PLUS is a trademark of Carpenter
Technology Corporation

(4)

FERRALIUM is a trademark of Langley Alloys, Ltd.

Zeron 100 is a trademark of Weir Material Services, Ltd.
* When the joint is fully heat treated after welding, Ni should
be 6.0-8.0%

(5)

the same value used for austenitic

stainless steel, 300F to 350F (150C
to 175C).

low heat input welding process, below

15 kJ/in. (0.6 kJ/mm), must be used, a
preheat of 200F-400F (95C-205C)
reduces rapid cooling and decreases
the amount of ferrite in the weld metal
and HAZ.

Preheat There is no need for

preheat on thicknesses 0.25 in. (6 mm)
and less on welds made with nickelenriched filler metals. In heavier sections and high restraint welds, preheat
may be used to advantage in minimizing
the chance of weld cracking. When a

Other stainless steels

Other types of stainless steels are
martensitic, ferritic and precipitation
32

three types follows and illustrates some

of the basic differences from the
austenitic alloys. Some of the more
common alloys and their welding filler
metals are shown in Table X.
Table X
Suggested filler metals for welding
some of the martensitic, ferritic and
precipitation hardening stainless
steels
Base
metal
AISI
(UNS)

Covered
welding
electrode
AWS A5.4
(UNS)

Bare welding
electrodes and
rods
AWS A5.9
(UNS)

Type 410 (wrought) E410

(S41000)
(W41010)

ER410
(W41040)

CA-15 (casting)
(J91540)

E410
(W41010)

ER410
(W41040)

CA-6NM (casting)
(J91540)

E410NiMo
(W41016)

ER410NiMo
(W41046)

Type 430 (wrought) E430

(S43000)
(W43010)
17-4PH
(S17400)
(1)

(1)

E630
(W37410)

ER430
(W43040)
(1)

ER630
(W37440)

When weld does not need to match bare metal

strength, E 308 (UNS W30810) or ER 308
(UNS W30840) are often used.

Martensitic stainless steels

The martensitic alloys can be hardened and strengthened by heat treatment and only slightly hardened by cold
working. They are strongly magnetic,
resist corrosion in mild environments
and have fairly good fabricating
qualities. The alloys are often selected
because of high mechanical properties
and low cost.
Weldability of the martensitic
stainlesses varies with alloy content,
particularly the amount of carbon. The
higher the carbon content, the greater
the need for preheat and postweld heat
treatment to produce sound welds.
While the wrought martensitic stainless
steels have limited use in process
industries, the cast grades have been

For the materials engineer

extensively used for heavy components
such as pump bowls, valve bodies and
compressor cases. CA-15 (UNS J91150)
was the standard alloy but has been
largely replaced by CA-6NM (UNS
J91540). Compared to CA-15, CA-6NM
has improved toughness and weldability,
along with better resistance to cavitation.
It is preferable to weld CA-6NM castings in the heat treated condition rather
than the as-cast. Welding is usually done
at room temperature although a preheat
of 250F to 300F (120C to 150C) may
be beneficial for large welds in heavy or
highly stressed sections. After welding,
the casting is heated to not higher than
1100F to 1150F (590C to 620C) and
air-cooled. When there is a special
hardness requirement, CA-6NM may be
given a normalizing heat treatment above
1750F (950C) and air cooled, followed
by a double temper of 1100F to 1150F
(590C to 620C). The casting should be
cooled to room temperature between
each tempering treatment.

Ferritic stainless steels

The ferritic alloys are not hardenable
by heat treatment and only slightly
hardenable by cold working. They are
magnetic and have good resistance to
corrosion in many environments. Most
typical of the ferritic stainless steels is
type 430 (UNS S43000), a straightchromium alloy with 16% to 18% chromium, 0.12% max. carbon, some minor
elements and the balance iron.
Weldability of the ferritic stainless
steels is generally better than the
martensitic. Exposure to high temperatures, such as in the weld heat affected
zone, causes a reduction in ductility and
toughness along with grain coarsening.
Solution annealing to prevent IGA is
done at 1450F (790C) for ferritic
stainless steels instead of 1900F to
1950F (1040C to 1065C) as for
austenitic stainless steels. There is
greater need for preheating and
postweld annealing as the thickness and
joint restraint increases.
33

For the materials engineer

Precipitation hardening
stainless steels

areas worn and corroded in later service

will sensitize the HAZ to IGA. Users can
and should request a 0.03% C maximum as an exception to the specification in order to prevent IGA in this alloy.
Normally, these castings are welded
without problems, but where welding is
extensive, special techniques may be
needed to prevent microfissures next to
the weld. Techniques available are low
interpass temperatures, low heat input
and peening of the weld to relieve
mechanical stresses.
The austenitic and duplex castings in
Table Vll are usually purchased to one
of the following specifications:

Iron-chromium-nickel alloys containing

precipitation-hardening elements such
as copper, aluminum and titanium have
good weldability, comparable to that of
the austenitic alloys, but are often used
for components which require little or
no welding. When welding is required, it
is best to weld these alloys in the
annealed condition prior to the final age
hardening heat treatment. These stainless steels are hardenable by a
combination of cold working and a lowtemperature heat treatment, 850F to
1100F (455C to 595C).

ASTM A 743 Castings, IronChromium, Iron-Chromium-Nickel,

Nickel-Base, Corrosion Resistant,
For General Application.

Corrosion resistant
stainless steel castings
Stainless steel castings are classified,
based on their end use, as corrosion
resistant or heat resistant, and are
designated accordingly by the first letter
C or H. The heat resistant grades are
generally higher in alloy content than
the corrosion types and in nearly all
cases have higher carbon. The following remarks apply to corrosion resistant
types and may not be applicable to the
heat resistant alloys. The chemical
composition of the more common
austenitic and duplex cast alloys is
shown in Table Vll (see page 25).
The most widely used, CF-3, CF-3M,
CF-8 and CF-8M grades, normally have
5% to 20% ferrite in the austenitic
matrix. The amount will vary with
composition, thermal history of the
casting and at different locations in the
casting. Ferrite is beneficial in minimizing casting cracks and improving
weldability. Some cast corrosion resistant stainless steel grades such as CN7M are fully austenitic by nature of their
composition. ASTM Specifications do
not as yet include the 0.03% carbon
grade for CN-7M as they do for the
standard grades. Weld repair of castings at the foundry or weld buildup of

ASTM A 744 Castings, IronChromium, Nickel-Base, Corrosion

Resistant, For Severe Service.
Both specifications require that the
casting be solution annealed which
largely removes alloy segregation and
dendritic structures occurring in castings, particularly in heavy sections. The
high temperature anneal of 1900F
(1040C) or higher, depending on the
alloy, promotes a more uniform chemical composition and microstructure, as
well as dissolving carbides. As a result
of the anneal, the casting is in the most
corrosion resistant state. For best
corrosion resistance, the widely used
CF-3M and CF-8M grades must be
annealed at 2050F (1120C), not
1900F (1040C) as allowed by the
specification.
Stainless steel castings are often
welded, either by fabricators making
assembly welds, during service life or
by the foundries weld repairing defects.
When the casting will be placed in a
severe corrosive environment, selecting
a low carbon version such as CF-3 or
CF-3M can avoid problems resulting
34

from the formation of chromium carbides in the weld HAZ. The same effect
of chromium carbides on IGA discussed
in wrought alloys is true for cast alloys.
The need for a low carbon version
applies not only for the initial fabrication
welds but also for later maintenance
overlay and weld buildup of cast components. When a low carbon grade is not
included in ASTM A743 or A744, an
exception to the specification can
usually be reached with the foundry.
One difference between A743 and
A744 is that A744 requires a full solution anneal after all weld repairs except
for minor repairs as defined in the
specification. Austenitic A743 castings
which are intended for general service
do not require the solution anneal to be
made after all weld repairs. Knowledge
of the intended service conditions is
helpful in selecting the correct material
specification and casting grade but if
this information is not available, a low
carbon grade of W744 is usually a good
choice.

Heat treatment of
stainless steel
Austenitic stainless steels, both wrought
forms and castings, are normally supplied in the solution annealed condition.
In solution annealing, the alloy is
heated to a high temperature, 1900F to
2150F (1040C to 1175C) depending
on the alloy type, and rapidly cooled,
usually by a water quench. At the
annealing temperature, chromium
carbides are put back into solution as
chromium and carbon, restoring the full
resistance to IGA of the alloy. The
anneal also removes the effect of cold
working and places the alloy in a soft,
ductile condition, however, the quenching operation may leave considerable
residual stresses.
In fabrication, even higher residual
stresses may be developed as a result
of forming operations and welding.

For the materials engineer

When weldments are solution annealed
and rapidly cooled, new residual
stresses are often introduced. These
stresses can cause movement after
machining, with the result that the part
exceeds dimensional tolerance limits.
Stress relieving of mild steel weldments
is frequently performed but it is best to
avoid stress relief treatments of stainless steel weldments unless absolutely
necessary. When it is necessary, two
alternates are available, namely:
Alternate 1 Use a low temperature
stress equalizing treatment at 600F to
800F (315C to 425C) with a hold of 4
hours per inch of thickness, followed by
a slow cool. Since the alloys have
excellent creep strength, the low temperature treatment removes only peak
stresses. The treatment is safe to use
with the standard grades such as 304
and 316 as well as the stabilized and
low carbon grades since the temperature is below that at which harmful
chromium carbides form.
Alternate 2 If the 600F to 800F
(315C to 425C) treatment is inadequate in reducing stresses to the level
desired, stress relieving in the range of
800F to 1700F (425C to 925C) may
be required. The higher the temperature
and the longer the time, the more
complete the stress relief. For example,
one hour at 1600F (870C) removes
about 85% of the residual stresses.
However, the standard grades such as
Types 304 and 316 cannot be heated in
this range without sacrificing corrosion
resistance as a result of carbide precipitation. A stabilized grade, e.g., 321, 347
or 348 or a low carbon grade, e.g.,
304L, 316L etc. should be used when
stress relief in this temperature range is
required. Refer to Figure 11, page 26,
in developing stress relief treatments
that will avoid carbide precipitation and
IGA in Type 304.

35

For the materials engineer

Table XI
Stainless steel products forms
Diameter
or size

Item

Description

Thickness

Width

Sheet

Coils and cut lengths:

Mill finishes Nos. 1, 2D & 2B
Pol. finishes Nos. 3, 4, 6, 7 & 8

under .187 in.

under .187 in.

24 in. & over

all widths

Strip

Cold finished, coils or cut lengths

under .187 in.

under 24 in.

Plates

Flat rolled or forged

.187 in. & over

over 10 in.

Bars

Hot finished rounds, squares,

octagons and hexagons
Hot finished flats

.125 in. & over

.25 in. to 10 in.

incl.

.25 in. & over

.375 in. & over

over 1.2 in.

Cold finished rounds, squares,

octagons and hexagons
Cold finished flats

Rods

Hot rolled rounds, squares,

octagons, and hexagons in coils
for cold rolling or cold drawing.

Wire

Cold finished only:

round, square, octagon,
hexagon, flat wire

Extrusions

Not considered standard shapes, but of wide interest. Currently limited in size to approximately 6.5 inches diameter circle, or structurals to 5 inches diameter.

Tubing

Dimensions by the outside diameter and wall thickness.

Piping

Nominal pipe size is the inside diameter from .125 inch to and including 12 inches. Nominal
pipe size is the outside diameter for 14 inches and larger diameters. The wall thickness is
dimensioned by schedules (5S, 10S, 20, 30, 40, 80, 120, 160, XX and variations thereof) for
all nominal pipe sizes.

0.010 in. to
under.187 in.

Material procurement
and storage guides

.062 in. to
under.375 in.

.25 in. to .75 in.

.5 in. & under

are included in weldments, it is useful

to review the extent to which the heat
affected zone of welds may suffer IGA
in the process fluid of interest.

Table XI shows the principal stainless

steel wrought product forms plate,
sheet, pipe and tubing most commonly used in welded fabrication. These
are normally purchased in the low
carbon modification for each grade,
although the alloys stabilized with
titanium or niobium-tantalum are also
used as discussed in the section on
welding for corrosion resistance. Bars
and structural shapes are sometimes
included in the fabrication. With bars
and shapes, the higher carbon grades
are normally used because of their
somewhat higher strength. When they

Surface finishes
Table Xll shows the standard finishes
for sheet and strip. The most widely
used finish for sheet is 2B. Polished
finishes are also available but are not
normally used in welded fabrications for
the chemical and other process
industries except food and medical
equipment.
There is no standard surface finish for
plate as there is for sheet and strip.
Plate is normally hot rolled, annealed
and pickled. Surface defects and
36

For the materials engineer

Table XII
Standard mechanical sheet finishes
Unpolished or rolled finishes:
No. 1
A rough, dull surface which results from hot rolling to the specified thickness followed by
annealing and descaling.
No. 2D

A dull finish which results from cold rolling followed by annealing and descaling, and may
perhaps get a final light roll pass through unpolished rolls. A 2D finish is used where
appearance is of no concern.

No. 2B

A bright, cold-rolled finish resulting in the same manner as No. 2D finish, except that the
annealed and descaled sheet receives a final light roll pass through polished rolls. This is the
general-purpose cold-rolled finish that can be used as is, or as a prliminary step to polishing.

No. 2BA or BA

Non-standard but widely offered bright annealed finish, highly reflective surface.

Polished finishes:
No. 3
An intermediate polished surface obtained by finishing with a 100-grit abrasive. Generally
used where a semi-finished polished surface is required. A No. 3 finish usually receives
additional polishing during fabrication.
No. 4

A polished surface obtained by finishing with a 120 150 mesh abrasive, following initial
grinding with coarser abrasives. This is a general-purpose bright finish with a visible grain
which presents mirror reflection.

No. 6

A dull satin finish having lower reflectivity than No. 4 finish. It is produced by Tampico
brushing the No. 4 finish in a medium of abrasive and oil. It is used for architectural
and ornamental applications where a high luster is undesirable, and to contrast with brighter
finishes.

No. 7

A highly reflective finish that is obtained by buffing finely ground surfaces but not to the
extent of removing the grit lines. It is used cheifly for architecural and ornamental purposes.

No. 8

The most reflective surface, which is obtained by polishing with successively finer abrasives
and buffing extensively until all grit lines from peliminary grinding operations are removed. It
is used for applications such as mirrors and reflectors.

Standard mechanical strip finishes:

No. 1
Approximates a 2D finish for sheet.
No. 2

Approximates a 2B finish for sheet:

BA

Bright annealed, highly reflective finish. Used extensively for automotive trim.

Mill-buffed

No. 2 or BA strip finish followed by buffing to produce a uniform colour and uniform reflectivity.
Used for automotive trim, household hardware and for subsequent chromium plating.

For more information consult: NiDI 9012, Finishes for stainless steels.

roughness in plate can initiate crevice

attack in severe environments. For such
services, it is necessary to negotiate
surface finish required with the
producer.
Pipe is not normally furnished to a
specific finish. Welded pipe is made
from cold finished coils in sizes up to

about 8 in. (200mm), and from sheet in

larger sizes. The finish on welded pipe
normally approaches the 2B or 2D finish
on sheet, except in the area of the weld.
The finish on extruded seamless pipe is
not quite as smooth but is normally
satisfactory from the standpoint of
corrosion.

37

For the materials engineer

Electropolishing is an electrochemical
process which provides a high luster
finish and is finding increased use in
applications where cleanability is a major
concern, such as bioprocessing and
paper mill head box equipment. The
process may be described as the reverse of electroplating in that there is an
electrolyte but the current is reversed
and metal is removed rather than plating
a new layer. The electropolishing process selectively reduces the peaks and
sharp edges that exist on the metal
surfaces, which in turn lessens the
chance of product build up and eases
cleaning. There is evidence that the
corrosion resistance is improved over
that of mechanically polished surfaces.
Electropolishing may be performed on
the completed fabrication rather than
sheet, strip or other starting products.
The surface roughness, that is the
distance between the peaks and valleys,
is reduced about 25% through
electropolishing. The surface may be
ground to a 180 to 250 grit finish before
electropolishing but mechanical polishing
on electropolished surfaces is avoided.

Purchasing guidelines

welded and removed from the rest of

the surface just prior to final cleaning
and inspection.
4. Pipe is normally ordered to either
ASTM A 312, which requires a final
heat treatment after welding, or to
ASTM A 774, which does not. Pipe
to A 312 is standard for most warehouse stock. Only five of the more
common grades are covered in A
778. A 778 tends to be used for the
larger diameter sizes where the low
carbon grades have proven to have
adequate corrosion resistance in the
as-welded condition. ASTM A 403
and A 774 are the comparable
specifications for stainless steel
fittings. Large diameter welded
stainless steel pipe spiral can be
obtained to ASTM A 409.
5. The interior finish in the area of the
weld is often of concern. Most
welded pipe producers achieve a
good finish in the area of the weld
but the finish desired should be
identified in procurement to avoid
misunderstandings.
6. Standard pipe lengths are 20 ft.
(6m), but longer lengths up to 60 ft.
(18m) are available. For smaller
diameter lines, considerable savings
can be made by ordering in longer
lengths and utilizing bends in lieu of
fittings.

The following guidelines are offered

when purchasing stainless steel to be
used in the fabrication of corrosion
resistant equipment.
1. Select the low carbon grades or the
stabilized alternates for welded
fabrications that will not be solution
annealed after fabrication.

Standard specifications for stainless

steel product forms for welded fabrication are shown in APPENDIX A. The
ASTM specifications do not cover
assembly welds such as required to
fabricate pipe assemblies, tanks and
other process equipment. Specfications
for fabrication weld quality are the
responsibility of the user and must be
included in procurement documents.

2. Specify a 2B finish for sheet.

Specify the finish the application
requires for plate in the
procurement documents.
3. Specify protective paper for sheet
and plate to be applied at the mill
when special surface protection
during storage and fabrication warrants. The protective paper can be
stripped back in the area to be

38

Part III
For the design engineer

Design for corrosion

services
Much can be done in the detailed design
to improve corrosion resistance and
obtain better service from less expensive grades.
There are two cardinal rules:
1. DESIGN FOR COMPLETE AND
FREE DRAINAGE.
2. ELIMINATE OR SEAL WELD
CREVICES.
Tank bottoms Figures 14-1 through
14-6 show six common tank bottom
arrangements. The square corner flat
bottom arrangement, Figure 14-1, invites
early failure from the inside at the corner
weld where sediment will collect, increasing the probability of crevice attack.
Moisture penetrating the flat bottom to
pad support invites rapid crevice corrosion from the underside.
The rounded bottom shown in Figure
14-2 is much more resistant from the
inside, but is actually worse from the
outside as condensation is funnelled
directly into the crevice between the tank
bottom and pad support. The grout used
to divert such condensation, Figure 14-3,
does help initially but soon shrinks back
and becomes a maintenance demand
itself. The drip skirt shown in Figure 14-4
is much the best arrangement for flat
bottom tanks. The concave bottom and
the dished head bottom on supports,
Figures 14-5 and 14-6, are very good
and superior to all flat bottom tanks not
only in corrosion resistance but also in
fatigue. Fatigue stresses from filling and
emptying are seldom considered in

design, but can be significant and have

led to failures in flat bottom tanks. The
concave and dished head arrangements
can withstand much greater fatigue
loadings than can flat bottoms.

Figure 14-1 Flat

bottom, square
corners worst.

Figure 14-2 Flat

bottom, rounded
corners good
corners, poor
outside.

Figure 14-3 Flat

bottom, rounded
corners, grouted
poor inside, poor
outside.

Figure 14-4 Flat

bottom, rounded
corners, drip skirt
good inside,
good outside.

Figure 14-5
Concave bottom,
rounded corners
good inside, good
outside, fatigue
resistant.

Figure 14-6
Dished head best
inside, best outside, fatigue
resistant.

39

For the design engineer

Tank bottom outlets Water left
standing in the bottom of stainless steel
tanks has been a source of tank bottom
failures in both fresh and saline waters.
Side outlets and centre outlets, shown
in Figures 14-7 and 14-8, allow for
convenient construction but invite early
failure of stainless steel tank bottoms.
Not only is a layer of stagnant water
held on the tank bottom but sediment
cannot be easily flushed out. A flush
side outlet and a recessed bottom outlet
as in Figures 14-9 and 14-10 allow the
bottom to be completely drained and all
debris and sediment to be flushed out,
leaving the bottom clean and dry. The
sloped arrangements shown in Figures
14-11 and 14-12 make it easier to flush
out and clean.

Bottom corner welds When the side

wall forms a right angle with the bottom,
the fillet weld is seldom as smooth as
shown in Figure 14-13. It is usually
rough and frequently varies in width
compensating for variations in fit up.
Because of the location, it is very difficult
to grind and blend the weld into the
adjacent sides. Debris tends to collect
and is difficult to remove, leading to
under sediment type crevice attack.
Unless welded from the outside as in
Figure 14-14, the crevice is vulnerable to
crevice attack. Rounding the corner and
moving the weld to the side wall overcomes both shortcomings as shown in
Figure 14-15. This construction has
much improved corrosion resistance and
also has better fatigue resistance.

Figure 14-7 Side

outlet, above
bottom poor.

Figure 14-8 Centre Figure 14-13

outlet, above
Corner weld from
bottom poor.
inside poor
inside, worst
outside.

Figure 14-9 Side

outlet, flush
good.

Figure 14-10
Centre outlet,
recessed good.

Figure 14-11 Side

outlet, flush,
sloped best.

Figure 14-14
Corner weld from
both sides poor
inside, good
outside.

Figure 14-15 Side wall in lieu of corner

weld best inside, good outside, fatigue
resistant.
Attachments and structurals All
attachments create potential crevice
sites. Figure 14-16 shows a tray support
angle with intermittent welds adequate
for strength. There is a severe crevice
between the angle and the inside wall of
the vessel which will become filled with
debris and invite premature failure from

Figure 14-12
Centre outlet,
recessed, sloped
best.
40

For the design engineer

crevice corrosion.
Figure 14-17 shows
the same tray
support with a
continuous seal weld
at the top preventing
unwanted material
from finding its way
Figure 14-16 Tray
down the wall and
into the crevice. The support, staggered
strength weld
angle to side wall
crevice is still open adequate support,
from the bottom but severe crevice.
this is a much less severe crevice which
vapors but not material can still enter.
Figure 14-18 shows a full seal weld at
the top and bottom of the tray support
angle. Here the crevice is fully sealed.

always be done. It is good practice to

drill a weep hole through the outer wall
and must be done if the vessel is to
receive a stress relief or solution anneal,
otherwise the expansion of the trapped
air could damage the vessel wall.
Figure 14-21 shows structural angles
positioned so they can drain, an important factor when shutting down and
flushing out. Angles should never be
positioned as in the top section of Figure
14-22. The best position for complete
drainage is shown in the lower view.
When channels are used, drain holes

Figure 14-21
Position of angles.

Figure 14-17 Tray

support, full seal,
weld top good
crevice resistance.

Figure 14-18 Tray

support, full seal
weld top & bottom
best crevice
resistance.

Figure 14-19
Reinforcing pad,
staggered welds
adequate strength.

Figure 14-20
Reinforced pad,
seal weld best
crevice resistance.

Figure 14-19 shows a reinforcing pad

to which other attachments are frequently welded. The intermittent weld
creates a severe pad-to-sidewall crevice
inviting premature failure. It takes very
little more time to complete the seal weld
as shown in Figure 14-20 which should

Figure 14-22
Position of angles.

should be drilled about every 12 in.

(300 mm) in the centre, unless they can
be positioned as in the right hand view
of Figure 14-23.
Continuous fillet welds on support
beams will seal the severe beam to
horizontal plate crevice shown in the
upper section of Figure 14-24.
Baffles in tanks and heat exchangers
create dead spaces where sediment

Figure 14-23 Position of channels.

Figure 14-24 Vertical beams.

41

For the design engineer

can collect and where full cleaning is
difficult. Figure 14-25 shows a cut out at
the lower corner of a tank baffle and
Figure 14-26 a cut out in the lower
portion of a heat exchanger tube support plate. Both arrangements reduce
debris collection and facilitate cleaning.
Heaters and inlets Heaters should
be located so they do not cause hot
Figure 14-28 Poor and good designs
for mixing concentrated and dilute
solutions.
stainless steel pipe rather than butt
weld, as in Figure 14-29. On larger
diameter pipe, over 2 in. (50 mm), a
Figure 14-25
Figure 14-26 Heat backing ring as shown in Figure 14-30 is
Corner baffle cut
exchanger, baffle
often used. Both designs may be
out good.
cut out good.
satisfactory for those services where the
spots on the vessel wall. In Figure 14-27, stainless steel alloy has adequate
crevice corrosion resistance. Because of
the poor location of heaters creates hot
the crevice formed, these designs often
spots which in turn may result in higher
corrosion in the area between the heater lead to unnecessary corrosion in
and vessel wall. The good design avoids aggressive environments and are not
hot spots by centrally locating the heater. recommended. The backing ring has the
further disadvantage of protruding into
When a concentrated solution is
the flow stream, which in turn can cause
added to a vessel, it should not be
product build up or unnecessary
introduced at the side as shown in the
turbulence.
poor design of Figure 14-28. Side
Very often stainless steel piping is
introduction causes concentration and
uneven mixing at the side wall. With the installed as commercial quality, that is
good design, mixing takes place away
from the side wall. It is also good
design practice to introduce feed below
the liquid level to avoid splashing and
drying above the liquid line.
Pipe welds It is frequently convenient to socket weld small diameter

Figure 14-29
Socket weld joint
severe crevice.

Figure 14-30 Pipe

weld made with
backing ring
severe crevice.

without imposing code standards such

as the ASME or the American Petroleum Institute, API, which require full
penetration butt welds. When procurement does not require full penetration
smooth ID butt welds, it is all too common to have a beautiful looking weld
from the outside but incomplete pen-

Figure 14-27 Poor and good designs

for the location of heaters in a vessel.
42

For the design engineer

The hand fed filler metal method is
more widely used in the chemical
process industry but the experience of
the particular company or welders,
strongly influences the selection. It is
important that the root bead have
adequate and uniform amounts of filler
metal melted into the weld for best
corrosion resistance. This addition is
readily obtained with consumable
inserts or by skilled welders using the
hand fed filler metal method.
There are a number of automatic
GTAW machines available for root
pass and fill welding. The root pass
can be made using an insert, with
automatic wire feed or in thin wall pipe,
single pass welds can be made without
filler metal addition. The ID root
contour of automatic welds is very
consistent and it is an excellent
process to use where the economics
are favourable. Automatic GTAW is a
particular advantage for tubing and
pipe 2 in. (50 mm) diameter and less.
Three good pipe-to-flange welding
arrangements are shown in Figures
14-33, 14-34 and 14-35. The recessed
arrangement shown in 14-33 avoids
the need for machining or grinding
smooth the surface of the weld on the
flange face in Figure 14-34. Both these
arrangements are suitable when the
flange is of the same material as the
pipe. Neither is suitable when carbon
steel or ductile iron flanges are used
on stainless steel pipe. In this case a
stub end arrangement shown in Figure
14-35 is preferred. In the case of pressure

etration on the
inside, such as
shown in Figure
14-31. Many
corrosion failures
originate in crevices created by
incomplete penetration at the root Figure 14-31 Pipe
weld with incomof pipe butt welds.
plete penetration
Since ASTM does
severe crevice.
not cover fabrication, procurement specifications must
specify full penetration and smooth ID
for the root bead of butt welds when the
weld quality is not covered by other
specifications.
The preferred pipe butt welding
procedure to insure high quality root
welds is the use of GTAW for the root
pass with an inert gas backing. In
manual root pass welds, the hand fed
filler metal technique or the use of
consumable inserts is commonly used.
Figure 14-32 shows some standard
consumable insert designs. Properly
made welds with either technique can
provide a crevice free ID surface with
minimum bead convexity or concavity.

Figure 14-32 Standard consumable

inserts, (from AWS A5.30).

Figure 14-33 Pipe

recessed flange
and pipe, same
alloy good.
43

Figure 14-34 Pipe

flush, pipe and
flange same alloy
better.

For the design engineer

piping, the flange
design must also
be in accordance
with the applicable
design or fabrication specification.
For piping and
heat exchanger
Figure 14-35 Stub
tubing to drain
end, flange carbon
completely, it is
steel or ductile iron
necessary to slope very good.
the piping or heat
exchanger just enough so that water will
drain and not be trapped where the pipe
or tubing sags slightly between support
points. Figure 14-36 shows how a water
film tends to remain in horizontal runs of
pipe or tubing and how water drains
when sloped.

Figure 14-36
(A) Horizontal (standard) poor.
(B) Horizontal sloped very good.

44

Appendix A
Specifications for stainless steel for welded fabrication
ASTM A240 Heat resisting chromium and chromium-nickel stainless
steel plate, sheet and strip for pressure vessels.
A240 is the basic specification for
procurement of stainless steel for
welded fabrication. A240 requires
solution annealing at the mill. This
specification includes 40 austenitic, 4
duplex and 16 ferritic grades. Caution:
Care must be taken to select the low
carbon or stabilized grades for
corrosion resistant services, as the
higher carbon grades, used primarily in
heat resistant applications, are included.
ASTM A262 Detecting
susceptibility to intergranular attack
in austenitic stainless steels.
A262 is a supplemental specification
that covers five tests that can be included in procurement specifications
when maximum resistance to
intergranular attack is required. When
A262 is used, the criteria to be met in
the test (practice) must be included as
pass/fail criteria are not part of A262.
ASTM A264 - Stainless chromiumnickel steel-clad plate, sheet and
strip.
A264 is the specification for the clad
construction using the austenitic grades
covered in A240.
ASTM A265 - Nickel and nickel-base
alloy-clad steel plate.
A265 is the specification for clad construction using the ten highly alloyed
nickel and nickel-base grades covered
under separately identified ASTM B
section specifications.

ASTM A312 Seamless and welded

austenitic stainless steel pipe.
ASTM A403 Wrought austenitic
stainless steel piping fittings.
A312 (pipe) and A403 (fittings) are the
older specifications for welded
austenitic stainless steel pipe for
aggressive environments developed
for, and widely used by, the chemical
industry. Both products require a
solution anneal after welding. Most of
the common stainless steels are
covered. Caution: Care must be taken
to select the low carbon or stabilized
grades for corrosion resistant services,
as the higher carbon grades are also
included in A 312. Sizes from 1/8 inch
to 30 inch diameter are included.
ASTM A778 Welded, unannealed
austenitic stainless steel tubular
products.
ASTM A774 As-welded austenitic
stainless steel fittings for general
corrosive services at low and moderated temperatures.
A778 (pipe) and A774 (fittings) are used
where the low carbon and stabilized
grades can be used in the as-welded
condition. Solution annealing after
welding is not required. Only low carbon
and stabilized grades are included in
these specifications. Sizes from 3
inches to 48 inches are covered.
ASTM A409 - Welded large diameter
austenitic steel pipe for corrosive or
high temperature service.
A409 covers light wall, spiral welded, as
well as straight seam welded, pipe 14
inches to 30 inches in diameter.

45

Additional requirements:
Few ASTM pipe and fitting specifications require pickling after production.

2. Inert gas backup on the inside of the

pipe during welding to minimize
oxidation and heat tint.

ASTM specifications do not cover shop

fabrication of pipe and fittings. The user
must develop his own specifications for
butt welding and fabrication to pipe
drawings. Important points to include
are the following:

3. Matching composition or higher Mo

content filler metal for Mo-containing
grades.
4. Protection of piping with protective
end caps to minimize contamination
during shipment and storage.

1. Full penetration, smooth ID, TIG

made root beads.

Bibliography
ANSI/AWS D10.4 86, Recommended
Practices for Welding Austenitic Chromium-Nickel Stainless Steel Piping and
Tubing.

American Welding Society, Welding

Handbook, Volume 4, Seventh Edition.

ANSI/AWS D10.11 87, Recommended

Practices for Root Pass Welding of Pipe
Without Backing.
ASTM A380, Standard Recommended
Practice for Cleaning and Descaling
Stainless Steel Parts, Equipment, and
Systems.

ASM International, Metals Handbook,

Ninth Edition, Volume 6, Welding,
Brazing and Soldering.
AWS B2.1 84, Standard for Welding
Procedure and Performance Qualification.

ASME Boiler and Pressure Vessel

Code, Section Nine.
Fabrication and Post-Fabrication
Cleanup of Stainless Steels, by Arthur
H. Tuthill, NiDI Technical Series
No 10 004.

American Iron and Steel Institute,

Welding of Stainless Steels and Other
Joining Methods, distributed by Nickel
Development Institute, Publication
o
N 9002.

Acknowledgement
The authors are indebted to Richard B.
Hitchcock and David E. Jordan for their

valuable technical comments.

46