Residual stress and distortion

The solidification and contraction of the weld bead

will induce strain and consequently stresses in the

joint. In the case of the contraction of the solid

metal the stress will be equal to the change in temperature

as the metal cools from its melting point

to its ambient temperature. Under normal circumstances

movement of the weld bead is restricted by

the adjacent body panels in the vehicle structure,

and the stress which could be generated is given by

the Young’s modulus of the material. This stress

level often exceeds the elastic limit of the material

Table 12.6MIG/MAG recommended operating conditions

Metal Wire feed Welding Gas

thickness Shielding Wire dia. Current Voltage speed speed flow

Application (mm) gas (mm) (A) (V) (m/min) (mm/min) (l/min)

Horizontal/vertical fillet 1.0 Argoshield 5 0.8 66 14 3.4 900 12

1.6 Argoshield 5 0.8 130 18 9.7 600 14

Argoshield 5 1.0 160 17 5.9 563 14

3.0 Argoshield TC 1.0 180 20 6.4 391 15

Argoshield TC 1.2 180 17 3.8 430 15

6.0 Argoshield TC 1.2 260 27 7.0 480 16

10.0 Argoshield TC 1.2 290 27 7.4 430 16

Butt weld, flat position 1.0 Argoshield 5 0.8 66 14 3.0 750 12

1.6 Argoshield 5 0.8 84 18 4.0 800 14

Argoshield 5 1.0 125 16 4.0 640 14

3.0 Argoshield TC 1.0 140 19 4.0 350 15

6.0 Argoshield TC 1.2 300 28 7.6 650 16

10.0 Argoshield TC 1.2 300 28 7.6 680 16

320 28 7.6 580 16

Butt weld, overhead 1.0 Argoshield 5 0.8 50 15 4.0 360 12

1.6 Argoshield 5 1.0 150 17 5.9 600 14

3.0 Argoshield TC 1.2 180 19 4.3 520 15

Butt weld, vertical up 1.0 Argoshield 5 0.8 45 15 3.5 350 12

1.6 Argoshield 5 1.0 130 17 4.7 410 14

3.0 Argoshield TC 1.2 135 17 3.1 275 14

336Repair of Vehicle Bodies

and plastic deformation takes place. The stresses

locked in the material, which may reach levels up to

the elastic limit of the material, are called residual

stresses, and the deformation of the material is

known as distortion.

Residual stresses

In a butt weld the weld bead will tend to contract

longitudinally and transversely, and this will induce

tensile stresses in the weld and also balancing compressive

stresses in the sheet material (Figure 12.34).

In the joining processes which rely on heating or

fusion, it is difficult to prevent the formation of residual

stresses. If the heat affected zone is ductile and

defect free (as in thin-sheet panel steel) the presence

of some residual stresses may be acceptable. The

most common technique used to relieve residual

stress in thicker materials is post-weld heat treatment.

This consists of uniform heating of the joint to

a temperature at which the yield stress of the material

is lowered and the residual stresses are relieved

by plastic deformation.

(Figure 12.35). Distortion may result in unacceptable

appearance (buckled body panels), prevention

of subassembly fabrication or the inability of the

structure to perform its intended function (alignment

of body panels after welding).

Figure 12.34Residual stress in a simple butt weld

(BOC Ltd)

Figure 12.35Types of distortion: (a) longitudinal

and transverse shrinkage (b) angular distortion (c)

out-of-plane buckling distortion (BOC Ltd)

Distortion

Distortion may take the form of a change in dimensions

of the joint, transverse or longitudinal shrinkage,

angular movement, or out-of-plane buckling

Gas shielded arc welding 337

The following steps may be taken to minimize

distortion:

1 Use the minimum amount of weld metal. Overwelding

and excessive reinforcement should be

avoided in fillet welds and flat butt welds.

Intermittent or stitch welding may be used.

2 Use square edge on narrow gap procedures to

reduce angular distortion when welding.

3 Use high travel speed and low heat input to

limit heat build-up in the panels to be welded.

4 Use the backstep weld sequence or preset the

joint.

Incorrect weld geometry and

Appearance, defects, loss of

Properties

Overfill

Overfill or excessive reinforcement may be described

as the presence of weld metal which exceeds that

required for the joint (see Figure 12.36). This creates

an unacceptable appearance or surface finish and

would require weld dressing.

Root concavity or suck-back

This may occur if the rear of the weld pool is too hot

or large, and the combined effect of contraction and

surface tension results in a root surface which is

concave (Figure 12.38). The surface profile will

often be smooth and less likely to produce stress than

over-penetration or lack of penetration; however, it

may result in unacceptable loss of cross-section. The

cause of this defect is incorrect parameter selection,

in particular the use of high current or slow travel

speed.

Figure 12.36Excessive reinforcement in (a) butt weld

and (b) fillet weld (BOC Ltd )

Figure 12.37Root convexity: excessive penetration

(BOC Ltd )

Figure 12.38Root concavity: suck-back (BOC Ltd )

(a)

(b)

(c)

Figure 12.39(a) Underfill and (b), (c) undercut

(BOC Ltd )

Root convexity

This results from excessive penetration on the underside

of a full penetration weld. The cause of this

defect is incorrect welding parameter selection

(shielding gas or operating technique) (Figure 12.37).

Underfill and undercut

If insufficient weld metal is deposited in the joint,

the parent material may remain unfused and the joint

may be underfilled (Figure 12.39a). Alternatively,

the weld area may be melted but the combined effect

of the arc force and the flow of the molten metal

may prevent complete joint filling or produce a

depression in the surface at the weld boundaries: this

is known as undercut (Figure 12.39b). Underfill is

usually avoided by careful placement and operating

technique. Undercut normally occurs at high speeds

and high current, and may be avoided by reducing

both current and speed.

338Repair of Vehicle Bodies

Lack of penetration

This is classed as partial penetration and is not acceptable,

particularly with respect to fatigue, loading and

corrosion resistance on the joint, where full penetration

is required. The defect results from poor joint

preparation and unsuitable welding operating parameters,

especially current settings (Figure 12.40).

The source of the gas may be a chemical reaction

such as the formation of carbon monoxide from

carbon and oxygen, or the expulsion of gases

which have been dissolved by the liquid metal but

are relatively insoluble in the solid phase. The

problem is best controlled by eliminating the

source of the pore generating gas. This may be

achieved by effective shielding and the prevention

of leaks, the ingress of air or moisture to gas lines,

and joint surface contamination from oil, grease,

moisture or paint.

Inclusions

Non-metallic inclusions may be trapped in the

joint owing to inadequate cleaning. Inclusions tend

to be more angular than gas pores and can result in

higher stress levels as well as forming a plane of

weakness in the weld. The cause of this defect is

usually insufficient inter-run cleaning, incorrect

welding parameters or incorrect gas shielding.

Solidification cracks (hot cracking)

Solidification cracking occurs when the solidifying

weld pool is exposed to transverse shrinkage

(Figure 12.42). The last part of the weld pool to

solidify is usually rich in low-melting-point impurity

elements, and liquid films are often present at

the centre line of the weld; as a result the ductility

of the weld metal is lowered. The transverse contraction

strain can be significant and, if the lowductility

area along the centre line of the weld is

unable to accommodate this strain, cracking takes

place. The problem is usually avoided by careful

selection of the weld metal composition.

Figure 12.40Lack of penetration (BOC Ltd)

(a)

(b)

Figure 12.41Lack of fusion (a) at sidewall and

(b) inter-run (BOC Ltd)

Figure 12.42Solidification cracks (BOC Ltd)

Lack of fusion

Lack of fusion between successive runs or between

the weld bead and the parent metal can produce serious

crack-like defects. These defects reduce the loadcarrying

area of the joint and produce stresses, and

are therefore unacceptable in welds which are subjected

to fatigue loading or static loading. The cause of

the defect is usually incorrect operating parameters,

high deposition rate, or low heat input (Figure 12.41).

Spatter

Spatter takes the form of small particles of metal

which are rejected from the arc or the weld pool

and adhere to the surface of the metal being joined.

Spatter may be unacceptable for purely cosmetic

reasons: it may also inhibit the application of surface

coatings (as in vehicle painting) and may initiate

corrosion. Spatter usually results from either

excessive voltage or insufficient induction in dip

transfer or argon enriched gas mixtures.

Porosity

Porosity results when bubbles of gas are nucleated

in the weld pool and trapped during solidification.

Gas shielded arc welding 339

12.15 Weld testing and inspection

Weld testing and inspection can be divided into

two types. In non-destructive testing, the test samples

are not destroyed in the process. In destructive

testing, the test samples are destroyed in the

process.

Visible examination

This is normally carried out during and after welding

and prior to any other non-destructive or

destructive test being used. This visual check will

usually determine the following:

Weld size

Profile or weld face shape

Surface defects in weld face

Undercut and overlap

Root defects

Weld penetration

Surface slag inclusions.

Non-destructive testing

Non-destructive testing of weld samples is normally

carried out using the following methods.

Macro examination

This method uses a low-power magnification to

examine weld specimens which have been levelled,

polished and etched to detect the following: lack of

fusion, lack of penetration, porosity, oxide inclusions,

internal cracks.

Crack detection

By using dye penetrant, surface defects in both ferrous

and non-ferrous metals may be detected. A

solution of coloured dye is sprayed on to the weld

and parent metal and allowed to soak. The dye is

then washed off and the surface dried. A liquid

developer is then sprayed on to the weld to give a

uniform dry powder coating which is white in

colour. The coloured dye oozes out of any crack in

the weld into the white coating and can be seen

in normal lighting conditions.

Magnetic particle method

Surface defects in mild steel and low-alloy steels

may be revealed by this method. The test specimen

is connected to a special power source. The magnetic

test liquid is sprayed along the weld. The

magnetic particles then collect along a line of

crack when the current flows through the weld.

Better test results are obtained if the surface of the

weld has first been ground smooth.

As these penetrants and magnetic test liquids

give off harmful vapours, the work must be carried

out under well ventilated conditions.

Ultrasonic inspection

This method uses sound waves which are passed

through the weld. They are transmitted as pulses

by a probe connected to an ultrasonic test set.

Defects reflect these pulses back to the probe

through a flexible cable which is attached to the

ultrasonic test set, where it is displayed on an

oscilloscope screen as deflections of a trace. These

deflections measure the location and size of the

defect. Grease is used between the probe and the

work to improve sound transmission. Also light

grinding of the weld or parent metal surface to

remove spatter may be required before testing.

Radiography

This method uses penetrating X-rays or gamma-rays

which are passed through the weld and recorded on

photographic film held in a light-proof container.

When the film has been processed, any defect in the

weld will show up as shadows in the photographic

image. Only trained and radiation classified personnel

may operate radiography equipment, as the

equipment is extremely dangerous.

Destructive testing

Mechanical testing is a destructive procedure and

therefore cannot be carried out on any component

required for use. Representative test samples produced

under similar conditions to the in-service

components, for example welding procedure tests,

are normally used, and accurate comparisons made.

The tests most frequently used to assess the

properties of welded joints are as follows.

Tensile test

This test can be used to assess the yield point, ultimate

tensile strength and elongation percentage of

the weld specimen. Failure usually occurs in the

parent material; therefore exact measurements are

not usually obtained for the weld itself, although in

this case the ultimate tensile strength (UTS) of the

340Repair of Vehicle Bodies

weld is higher than that of the parent metal. Test

specimens are cut from the designated area of the

weld assembly, the edges are smoothed and the

corners are radiused. If an elongation value is

required, then two centre punch marks often

50 mm apart are applied one on either side of the

weld; this is called a gauge length. The test equipment

can vary, but the basic principles are that two

sets of vice jaws are used to clamp the specimen,

hydraulic power is applied to force the jaws apart,

and a dial calibrated in tonnes or newtons records

the load. As the load increases, the dial registers

the amount of applied load until fracture occurs.

Bend test

Bend tests are carried out on butt welds and are

used to determine the soundness of the weld zone.

The test piece is the full thickness of the material,

with the weld bead reinforcement dressed flush on

both sides. It is common practice to take two specimens

from a test piece: one can be bent against the

face and the other bent against the root. A bend test

machine consists of a roller or former connected to

a ram and operated hydraulically. Unless otherwise

stated, the diameter of the former should be four

times the weld specimen thickness. The specimen

is placed on support rollers and the centre point of

the former is brought into contact with the weld

face or root. Pressure is then applied to bend the

weld specimen through either 90 or 180 degrees.

Nick-break test

In this test procedure the butt joint weld must first

be ground flush. Then two nicks (saw cuts) are

made, one on either end of the joint in the weld

specimen. The weld specimen may be broken during

bending, or alternatively the specimen may be

placed in the jaws of the tensile test machine and

struck with a hammer when tensioned.

Impact test

There are two main types of impact test, the

Charpy V notch and the Izod test. Both tests

employ a swinging pendulum which breaks accurately

prepared specimens to which a notch has

been applied. Both the tests have the same principle,

which is to determine the energy measured in

joules absorbed by the notched test piece at a specified

temperature as it is broken from a single

swinging pendulum. During fracture, energy from

the pendulum will be absorbed by the specimen,

more energy being absorbed by strong materials

than by brittle materials. The distance the pendulum

swings after fracture is measured by a pointer

on a dial calibrated in joules. The lower the value

indicated, the more brittle the specimen; conversely,

the higher the reading, the greater the

toughness.