MECHANICAL TESTING OF WELDED JOINTS

Mechanical testing is used to describe a group of test methods

for establishing or confirming the minimum level of mechanical
properties of a material or a completed weld. Most of these
tests destroy some part of the object being tested thus they are
called destructive tests.

Mechanical tests divide into qualitative and quantitative

methods

 Qualitative methods ( have units ):

tension test , hardness test , toughness test

 Quantitative methods ( have no units) :

Bend test , fracture test , macro test

 Destructive Tests :

- Macroscopic & microscopic examinations

- bending test
- Tension test
- Hardness test
- Charpy vee notch test
- Izod test
- crack tip open displacement test
- nick break test
- Chemical test

Page 1 of 41
a- Metallographic examinations
Are used to determine the following weld characteristics:
- Soundness of the joint
- Number of weld passes
- Extent of the heat affected zone
- Location and depth of the weld penetration
- Distribution of nonmetallic inclusions in the joint
- Metallurgical structure in the weld metal and the heat
affected zone

There are two types of metallographic examinations

1- Macroscopic Examination

Macroscopic Examination is a process used to reveal and

study the weld structure. it is polished or polished and etched
to bring out the structure, and examined at a low magnification
(up to 10 power).
The joint penetration and extent of dilution of the filler metal by
the base metal can be determined. It also reveals the presence
of slag, flux, porosity, and cracks in the weld and heat affected
zones. If macroscopic examination is used as an inspection
procedure, sampling should be done in the early stages of
manufacturing to allow for corrective action to be taken in the
event that it is necessary.
Specimens may be cut from the test assembly by thermal or
mechanical methods. If they are machined from the test
assembly using a lathe, milling machine or shaper the usual
procedure is to take a rough cut followed by a finishing cut
using sharp tools. Grinding operations are usually performed in
a similar fashion using free-cutting wheels and a light finishing
cut. If a finer detail is required then the specimen should be
polished with metallographic papers.

After the surface preparation is completed, the sample is

cleaned with a suitable solvent. All grease, oil and other
residue must be removed or it will produce uneven etching. It is
important to remember not to touch the sample after it has
been cleaned or this will contaminate the sample.

Page 2 of 41
ASTM E340, Standard Method for Macroetching Metals and Alloys
provides recommended solutions and procedures for macroetching.
When handling chemicals and mixing solutions caution must be
exercised.

Macro etched specimen

Etching solutions for macroscopic examination of ferrous

alloys :

Page 3 of 41
2-microscopic examination

Microscopic examination is a carefully polished and etched

process used when it is important to examine extremely small
discontinuities or metallurgical structures at a high
magnification (50 power and over).
a microspecimen obtained by sectioning test welds or
production welds. Thermal cutting, machining, or grinding is
then used Additional preparation may be required depending
on the desired surface needed. When thermal cutting
processes are used to extract specimens the entire heat
affected zone created must be removed before conducting any
metallographic examination.
Specimens must be polished and etched before being
examined under the microscope. This type of examination will
reveal the microstructure of the base metal, heat affected zone,
fusion zone, and weld metal.

Page 4 of 41
ASTM E3, Standard Methods of Preparation of Metallographic
Specimens gives procedures for the selection, cutting, mounting,
and polishing of metallographic specimens.
ASTM E407, Standard Method for Microetching Metals and Alloys
provides recommended chemical solutions for etching various
metals and alloys and safety precautions in the handling of etching
chemicals.
Etching solutions for microscopic examination of ferrous
alloys :

Page 5 of 41
Etching solutions for non-ferrous alloys :

Page 6 of 41
b- Bend Test

The bend test is a test method that is used to evaluate the

ductility and soundness of welded joints.

There are two different bend testing methods:

 guided bend test

 free bend test

1- Guided Bend Test

The guided bend test is commonly used in welder and

procedure qualification tests to determine the ability of the
welder to make sound welds. The test is performed by bending
prepared specimens of a specific dimension (usually specified
in the relevant code) in a special jig. The dimensions of the jig
will vary with specimen thickness and material.

Bend test jig dimensions

Page 7 of 41
 There are two kinds of bending jigs.
Plunger type bending jig
.uses a hydraulic ram to force the plunger into the die

Wrap-around bending jig

wrap-around type bending jig.

Page 8 of 41
the strain applied to the test specimen depends on the spacing
of the rollers and the radius of the male member. The strain on
the outside fibre of the bend specimen can be approximated
from the following formula:

 = 100 t / (2R + t)

Where
 = strain, %
t = bend test specimen thickness, mm. (in)
R = inside bend radius, mm. (in)

Note on the diagram that as the strength of the material

increases so does the bend radius, hence the applied strain is
reduced.

When preparing bend specimens prevent stress raisers that

lead to a premature failure of the specimen by radiusing all
edges and ensuring that all grinding marks are parallel to the
length of the specimen.

When performing qualification tests the specimen thickness

and bend radius are chosen according to the ductility of the
metal being tested. An elongation in the outside fibre of 20
percent can be easily achieved on sound mild steel welds.
Bend tests will consistently fail if the specimens contain weld
discontinuities that are on are near the surface of the material.

After bending, the welds are examined for the presence of

discontinuities. Many welding standards and specifications
consider that a bend specimen has failed if on examination of
the convex surface after bending there is a crack or open
defect exceeding 3mm (1/8 in.).

Page 9 of 41
 Successful bend test

Unsuccessful bend test

There are three types of guided bend tests:

- root bend tests (longitudinal or transverse)

- face bend tests (longitudinal or transverse)
- Side bend tests

Page 10 of 41
A root bend test puts the weld root in tension while a face bend
test puts the weld face in tension. Both types are generally
used on material thicknesses of 10 mm (3/8 in.) or less. When
the material thickness is greater than 10 mm (3/8 in.) side bend
test specimens are chosen . Side bend test strains the entire
weld cross section, and thus is especially useful for exposing
defects near the midthickness that might not contribute to
failure in a face or root bend test.

2 - Free Bend Test

This test bends the specimen without being constrained by a

bending or test jig. Prior to the test a prebending operation is
necessary as shown in the diagram. The final bend operation is
performed in a vice or press. It is important to note that the
specimen may have the tendency to pop out of the press
during the bending process.

Prebending of free bend test specimen

Page 11 of 41
 Final bending of free bend specimen

Prior to the test, gauge marks are scribed in the weld metal and
then the distance between them is measured. After the test the
distance between the gauge marks is measured again. The
difference between the two measurements is an indication of
the ductility of the weld metal. The formula used to measure the
percent elongation is:

Elongation in % = (( L2 - L1 ) / L1) x 100

where L1 is the original gauge length before bending and L 2 is

the gauge length after bending.

Page 12 of 41
Bend Test Limitations

Bend testing is sensitive to the relative strengths of the weld

metal, the heat-affected zone, and the base metal.

Many problems can develop in transverse bend tests such as

an overmatching weld strength may prevent the weld zone from
conforming exactly to the bending die radius, and thus may
force the base metal to deform to a smaller radius. This will not
produce the desired elongation in the weld. Alternatively, with
an under matching weld strength, the specimen may bend in
the weld to a radius smaller than the bending die. In this case
failure may result when the weld metal ductility is exceeded,
and not because the weld metal contained a defect.
These problems with weld strength mismatch can be avoided
by using longitudinal bend specimens which have the bend
axis perpendicular to the weld axis. In this case all zones of the
welded joint (weld, heat affected zone, and base metal ) are
strained equally and simultaneously. This test is usually used
for the evaluation of joints in dissimilar metals.

Weld discontinuities in longitudinal bend tests that are oriented

parallel to the weld axis such as incomplete fusion, inadequate
joint penetration, or undercut are only moderately strained and
may not cause failure.

Page 13 of 41
c- Tension tests
are performed for the following reasons:

- test results are used in selecting materials for engineering

applications
- tensile properties are frequently included in the material
specifications to ensure quality
- often tensile properties are measured during the development
of new materials and processes so that different materials and
processes can be compared.
- tensile properties are often used to predict the behaviour of a
material under different forms of loading, other than uniaxial
tension.

The strength and ductility of metals are obtained from a

uniaxial tension test . If the loading is continued the specimen
will eventually break.
Stress - strain curve

Page 14 of 41
 tension tests involve applying a load to the ends of a standard
test specimen and recording the point at which the specimen
fails by permanent shape change (yielding) and by fracture. A
number of mechanical properties can be determined from a
tension test, including the following which are of particular
significance in welding:

- yield strength ( the stress at which permanent deformation

occurs)
- ultimate strength (the highest stress the material is able to
withstand)
- breaking or fracture strength(the stress at which the material
fails by breaking)
- ductility (the percentage of elongation or reduction of area of
a defined segment of the specimen)

Two specific types of tension test specimens are used .

All Weld Metal Test (taken from the weld material only) ,
Reduced Section Tension Test(taken across the weld).

Tension testing machine

Page 15 of 41
1 - All Weld Metal Test

This test is used to determine the tensile properties of a

specimen that consists entirely of weld metal. The test
specimen is oriented parallel to the weld axis.

There are two reasons for performing an all weld metal test:

- to qualify a filler metal or

- determine the properties of the weld metal in a particular
weldment.

To qualify a filler metal the melting of the base metal is

minimized when making a test weld.

The following are typically properties that are measured and

reported in an all weld metal tension test:

- Tensile strength
- yield strength
- Elongation
- Reduction of area

All weld metal test specimen

Page 16 of 41
2 - Reduced Section Tension Test
Two types of reduced section tension tests are used to test the
strength of the welded joint as a whole, including the parent
metal.

2.1 - Transverse Weld Tension Test

Due to the inclusion of the base metal, heat affected zones, and
weld metal the test results for a welded joint as a whole is not
possible in the transverse direction. When a transverse
specimen is tested the area with the lowest strength will
elongate and break first. If the weld metal strength is higher
than the unaffected base metal then the base metal will fracture
first. Therefore, this test should not be used for quantitative
comparisons of weld metals.

Transverse weld tests are used most often to qualify welding

procedures, to verify that the welding procedure will produce
welds of equal or greater strength than the design strength
requirements. only the tensile strength and the location of the
fracture are measured and reported in this type of test.

Transverse weld tension test specimen

Page 17 of 41
2.2 - Longitudinal Weld Tension Test

For this test, the direction of loading is parallel to the weld axis.
It differs from the all weld metal test in that the specimens
contain the weld, heat affected zones, and base metal. It is
important that during testing all of these areas are strained
equally and simultaneously. As a low weld or heat affected
zone ductility may cause fracture to occur below the strength
level of the base metal, only the tensile strength is typically
measured and reported. The elongation of the joint is also
measured on occasion.

Before testing either of these test specimens the least width

and corresponding thickness of the reduced section are
measured to calculate the cross-section area. The ultimate
tensile strength is determined by dividing the maximum load
reached during the test by the cross-sectional area.

Longitudinal weld tension test

Page 18 of 41
Page 19 of 41
Page 20 of 41
Page 21 of 41
Page 22 of 41
Page 23 of 41
 d- HARDNESS TEST

Hardness described as the ability of a material to resist

permanent or plastic deformation, and measured by its
resistance to indentation by an indenter of a standard shape
and size.

The hardness test used for evaluating the properties of metals.

hardness test and the tension test measure the resistance of
a metal to plastic flow.
In general, an indenter is pressed into the surface of the metal
to be tested under a specific load for a definite time interval,
and a measurement is made of the size or depth of the
indentation.

Measurements of hardness can provide information about the

metallurgical changes caused by welding. For example, in alloy
steels a high hardness could indicate the presence of
untempered martensite in the weld heat-affected zone, while a
low hardness may indicate an over-tempered condition. In cold-
worked or age-hardened metal, welding may result in
significantly lower heat-affected zone hardness due to
recrystallization or over aging.

Hardness testing is divided into two categories:

macrohardness and microhardness. Macrohardness refers to
testing with applied loads on the indenter of more than 1 kg
and covers, for example, the testing of tools, dies, sheet and
plate material, castings, forgings, etc. As a rule of thumb, the
workpiece should not be thinner than 10 times the indentation
depth. Microhardness refers to testing with applied loads of 1
kg and less on very thin material (down to 0.0125mm (0.0005
in.)). This covers extremely small parts, plated surfaces, thin
superficially hardened parts, and individual constituents of
materials.
When performing a hardness test on the cross section of a
welded joint, the surface should be ground and polished and
also etched to show the heat affected zone. Hardness
indentations can be made on any area of the weld or base metal
.

Page 24 of 41
In many cases the hardness indentations are made at equal
intervals across the entire cross section of the weld joint.
Typical examples for a fillet and butt weld are shown.
Hardness scan - fillet welds

Hardness scan - butt weld

The hardness testing methods in use today for testing metals are:
- Brinell
- Rockwell
- Vickers
- Knoop

Page 25 of 41
1- Brinell test

The Brinell test consists of applying a constant load (force),

usually 500 to 3000 kg, on a 10 mm (0.4 in.) diameter hardened
steel, or tungsten carbide, ball to the flat surface of a work
piece.
This produces a large indentation in the material, typically 2 to
6mm (0.08 to 0.24) in diameter.
Typically, a 500 kg load is used for testing soft metals, such as
copper and aluminum alloys. On the other hand a 3000 kg load
is used for testing harder materials, such as steels and cast
irons.
After holding the load for a specified time (usually 10 to 15
seconds for hard materials and approximately 30 seconds for
soft materials) the indentation diameter is measured.
It is important to adhere to this time period to ensure that the
plastic flow of the metal has stopped.
Brinell indentation process

Page 26 of 41
To calculate the hardness value, the mean diameter of the
indentation is determined by taking two readings at right
angles to each other.

HB = L / ((  x D / 2 ) x ( D x SQRT(D 2 - d 2 )))

where
L = the load in kg
D = the ball diameter in mm
d = the diameter of the indentation in mm

The actual Brinell number is based on the surface area of the

indentation.
The diameter of the indentation is measured with microscope
to the nearest 0.01 mm (0.0004 in.).

To ensure an accurate result, indentations must not be made

too close to the edge of the workpiece as the lack of supporting
material will produce an abnormally large and unsymmetrical
indentation.
indentations must not be made too close to one another as the
material may be cold worked by the first indentation producing
an abnormally small indentation.
The distance between centres of adjacent indentations should
be at least three times the diameter of the indentation. This will
produce a Brinell number error of 1% or less.

Brinell hardness tester

Page 27 of 41
 Indentation measurement

The limitations and fundamentals of the Brinell hardness test

method are :

- indentations should never be made on a curved surface with

a radius of less than 25 mm(1 in.)
- spacing of indentations must follow the above rules
- the load should be applied so that the direction of loading is
perpendicular to the workpiece surface within 2 o
- the load must be applied steadily to avoid overloading
- the Brinell test is not applicable to testing thin sheet.

- The workpiece should not be thinner than 10 times the

indentation depth.

Page 28 of 41
 The indentation depth can be calculated using the following
formula:

Depth(mm) = L / ( x D x HB )

where
L = the load in kg
D = the ball diameter in mm
Hb = the Brinell hardness number

- the practical range of hardness is 16 HB minimum using a

500 kg load for soft material to 444HB (627 HB for a carbide
indenter) using a 3000 kg load for hard material

Page 29 of 41
2- Rockwell hardness test

The Rockwell hardness testing method is the most widely used

for the following reasons:

- the test is very simple to perform thus requiring less operator

skill
- it can be used on most metals and alloys by changing the
loads and indenters
- readings can be taken in very little time, usually just a few
seconds
- it does not require optical measurements since all readings
are read directly off of a dial

The main difference between the Rockwell test and the Brinell
test methods is that the Rockwell test measures the hardness
by the depth of indentation made by a constant load impressed
upon the indenter. The type of indenter is a diamond, ground to
a 120-degree cone with a spherical apex that has a 0.2mm
(0.008 in.) radius.

The use of the light or minor load has the following two
purposes:
- it eliminates the backlash of the applied load
- removes the surface roughness thus improving the accuracy
of the major load

After applying the minor load a reference or set position is

established on the instruments dial. The major load is then
applied to produce a further indentation of the material. The
hardness number is the difference between the two readings.
The entire process takes about 5 to 10 seconds to complete.

Page 30 of 41
 Principle of the Rockwell hardness test

For hardened steels and cemented carbide material a diamond

cone indenter is used whereas for soft material such as cast
irons and nonferrous metals, a hardened steel ball of 1.6, 3.2, 5,
and 12.5 mm (1/16, 1/8, 1/4, and 1/2 in.) is used .

there are two types of Rockwell testing methods, Rockwell and

Rockwell superficial. The testing methods are identical. For
Rockwell testing a minor load of 10 kg is used with a major load
of 60, 100, or 150 kg irrespective of the type of indenter used.
For the superficial Rockwell test method a minor load of 3 kg is
used with a major load of 15, 30, or 45 kg.

Page 31 of 41
The Rockwell hardness value always consists of a number and
several letters. Each major load and indenter combination has a
specific letter assigned to it. The first letter is H which stands
for hardness. This is followed by R for Rockwell and finally the
letter that indicates the scale used. For example, a value of 80
HRB indicates a Rockwell hardness number of 80 on the B
scale.

The Superficial Rockwell hardness number also consists of a

number and letter combination. The dial reading is followed by
HR as described above and another alphanumeric qualifier to
describe the load and the indenter combination used. The letter
N designates the diamond indenter and the letters T, W, X and
Y designate the use of a steel ball indenter. For instance a
value of 80 HR30N indicates a Superficial Rockwell hardness
number of 80 using a load of 30 kg with a diamond indenter (N).
The H and the R stand for the same descriptions as for the
Rockwell test method.
The relationship between the scale designation, indenter diameter
and major load is shown in the following table.

Rockwell hardness testing parameters - part 1

Page 32 of 41
Rockwell hardness testing parameters - part 2

Rockwell hardness testing parameters - part 3

Page 33 of 41
 Rockwell hardness testing parameters - part 4

Rockwell hardness tester

Page 34 of 41
3- Vickers hardness test

In principle, the Vickers hardness test is similar to the Brinell

hardness test method. They both use an indenter of a predefined
shape that is pressed into the material to be tested. The load is then
removed and a measurement of the indentation is taken. For the
Vickers test the diagonals of the resulting indentation are measured.
The resulting hardness number is calculated by dividing the load by
the surface area of the indentation. It has the same disadvantage as
the Brinell test in that the size of the indentation has to be
measured.

Vickers hardness test

Page 35 of 41
The indenter for the Vickers hardness test method uses a
diamond and is in the form of a square-based pyramid with an
angle of 136o between faces. Loads of 5, 10, 20, 30, 50, 100, and
.120 kg are used but in practice the 50 kg load is the maximum

Vickers hardness indenter

The Vickers hardness method is widely used for testing welds

because the indentation is small thus making it easy to perform
a complete survey of the weld cross section.

When performing a Vickers test the specimen is brought to

within 1 mm of the indenter and the loading mechanism is
tripped which automatically applies the load. The load dwells
for a short period of time and then is removed. The diagonals of
the pyramid are then measured with the aid of a microscope
which is part of the testing machine. The microscope is
equipped with knife edges which are brought up to the edges of
two diagonally opposite corners of the indentation. The
distance between the knife edges is the length of the diagonal.
The hardness number is calculated using the following
formula:

Page 36 of 41
 HV = 1.854 x P / d2

Where
HV = the Vickers hardness number
P = the applied load in kg
d = the mean diagonal of the indentation in mm

Tables are also available which make the calculation

unnecessary.

4- Knoop Microhardness test

Page 37 of 41
The hardness is determined by measuring the size of the
indentation with a microscope and then using its dimensions in
formulas or conversion tables. Microhardness testing is
capable of providing information on the hardness
characteristics of materials that cannot be obtained with
hardness tests such as the Brinell, and Rockwell methods.
Because of the required degree of precision for both equipment
and operation, microhardness testing is usually performed in a
laboratory. The following are typical applications of
microhardness testing:
- measuring the hardness close to the edges of parts
- measuring the hardness of bonding or plating surface layers
- measuring the hardness of parts that are too thin or small to
be measured by other hardness testing methods, for example,
foils and wire
- measuring individual grains and inclusions in a metal
A Knoop indenter is a rhombic-base pyramidal that produces a
rhombic-shaped indentation.
Knoop indenter

The Knoop hardness number is the ratio of the load applied to

the indenter, P (kg) to the projected area, A (mm):

HK = P / A = P / C x l2

where
HK = Knoop hardness number

Page 38 of 41
 P = the applied load in kg
A = is the unrecovered projected
area of indentation in mm
l = is the measured length of the
long diagonal in mm
C = is the indenter constant of 0.07028

Comparison of Vickers and Knoop indention patterns

A tester used both Knoop and Vickers microhardness

measurements is shown in the next photograph.
Microhardness tester

5- Microhardness Test

Page 39 of 41
 Microhardness usually refers to indentations that are made
with loads that do not exceed 1 kg and are usually within the
range of 100 to 500 g. However, the term is related to the size of
the indentation rather than the load applied. The Vickers and
Knoop test methods are used for microhardness testing.

 The following are some of the factors that must be

considered when selecting the appropriate hardness testing
method:

- size of the workpiece

- equipment availability
- surface condition of the workpiece
- degree of flatness of the workpiece
- effect of indentation marks
- hardness range of the material to be tested
- shape of the workpiece

The Brinell, Rockwell, and Vickers methods may be used on

ferrous and nonferrous metals over a wide range of hardness.
The Rockwell and Vickers methods can be applied to small
areas that include portions of welds, and heat affected zones.
The Brinell method is used for measuring hardness values over
larger areas such as the face of welds, and base metal.

The following precautions must also be adhered to once a test

method has been selected:
- use the correct load
- impact loading must be avoided
- ensure the indenter is in good condition
- ensure the anvil is in good condition
- take great care in measuring the size of the indentation
- consider the thickness of the specimen before
performing the test (remember the general rule discussed
previously - the thickness should not be less than ten times
the depth of the indentation)
- ensure that the specimen is firmly supported

Page 40 of 41
Page 41 of 41