Welded Joints


Many metals (and some plastics) can be welded, but it there are often restrictions on the type of welding process that can be used. The ideal of fusion welding is that the joint is no weaker, or in any way inferior, to the joined materials.

Where a fabrication contains a number of welds it is essential that the design minimises shrinkage strains or high residual stresses can be present in joints and components which can lead to premature failure.

In the case of low carbon steels, the joint can in some instances actually be stronger than the joined steels. It must be noted that in practice many welding operations introduce defects which result in the joint properties being less than the theoretical indicated. The most common problem with fusion welded steel joints is hydrogen embrittlement leading to cracking. This may originate from damp coatings on consumable welding electrodes - it is essential that coated electrodes are stored in dry conditions. The other point to note is that many welding operations are carried out by hand and in these cases the quality of the joint is heavily dependent upon operator skill.

The carbon content of a steel has a major influence on it's weldability. Low carbon mild steel can be welded without difficulty, but for carbon contents above 0.2% much greater care is needed and in the case of high strength steels intended for equipment where fabrication by welding is very common, the carbon content is often limited to a maximum of 0.15%.

Aluminium and many of its alloys can be welded, but there can be problems - such as the production of oxide films and weld metal porosity, which can be avoided by using inert gas shielded processes. It is often necessary to carefully match the electrode composition to the alloy being joined.

Strength of Fillet Welds

The first diagram shows the widely used transverse fillet weld. This is assumed to be of length 'l' perpendicular to the plane of the diagram. The size of the fillet is equal to the leg 'h' of the largest inscribed isosceles triangle.

The common method of analysis assumes the weld to be cut through at the narrowest point, the 'throat', see second diagram. Any outward 'bulge' due to weld metal build up is ignored. The stress, Sx, on the weld throat is resolved in the normal and shear directions, giving Sn and Ss.

The throat area: A = hlcos45 = 0.707hl

Sx = P / A = P/0.707hl

Resolving Sx into the normal and tangential components gives:

Sn = Sx cos 45 = P/hl and Ss = Sx cos 45 = P/hl

These stresses can be combined as shown in the Mohr's circle in diagram 3 to give:

S1 = (P/2hl) + ((P/2hl)2 + (P/hl)2)0.5 = 1.618P/hl and

Ssmax = ((P/2hl)2 + (P/hl)2)0.5 = 1.118P/hl

While the equations above give good agreement with test results, for design purposes it is common to use a more elementary approach and assume that the average working shear stress is equal to the load divided by the throat area: Ss = P/0.707hl = 1.414P/hl

Fatigue Strength Reduction Factors

The following fatigue strength reduction factors are suggested in 'Arc Welding in Design, Manufacturing and Construction' - James F. Lincoln Arc Welding Foundation, 1939:

Reinforced butt weld1.2
Toe of transverse fillet weld1.5
End of parallel fillet weld2.7
T-butt joint with sharp corners2

For a detailed discussion of the fatigue strength of welded structures see: 'Fatigue Strength of Welded Structures', by S J Maddox, Abington publishing, 1991. ISBN 1 85573 013 8.

Classification of Welding Electrodes

BS 639 uses a rather complicated combination of letters and numbers, some compulsory and some optional, to classify electrodes. The standard should be consulted for details.

The American Welding Society (AWS) Standard A5.1 uses a less complex notation starting with a prefix E followed by 2 sets of digits:

E: Electrode

xxx Two or three digits indicating the tensile strength in 10,000 psi units

xx Two digits indicating the coating type and application.

Return to Module Introduction

David Grieve 26th August 1999, revised 29th April 2003.