IFMASS8

354

capacities of the simulator as well as on the material, peak temperature of the thermal

cycle, and necessary size of the zone with simulated microstructure.

The lecture presents most important results of researches performed in the Research

and Development Institute of Metalna in Maribor, and in the Welding Laboratory of

Mechanical Engineering Faculty, Maribor, in the last 15 years, from since the thermal

cycle simulator was available. Some other results, obtained in the researches with our

participation are also included. Those results are used to explain applied approaches to

study fracture and fatigue properties of welded joints in HSLA structural steels.

1. WAYS TO TEST THE STRENGTH OF WELDED JOINT

Static and dynamic strength of welded structures with crack-like defects can be evalu-

ated using different types of precracked specimens. The reliability of these results,

obtained experimentally, depends mostly on the shape and size of specimens used.

In order to assess the actual strength of workshop-quality welded structures, full-size

models are loaded up to fracture, or to plastic yielding. These so-called proof tests of

full-scale samples are sometimes used in series production. Different types of proof

tests include strain- or displacement-measurement tests and fracture or burst tests.

Tests performed on structures with a crack prepared in the area of the lowest fracture

resistance provide reliable fracture-toughness data. The effect of structural shape and

the effects of residual stresses are included in experimental fracture-toughness data. In

the case of special-order production of huge welded structures (cranes, bridges, halls,

boilers, pressure vessels) full-scale tests are extremely expensive and are not often

used. Such well documented tests had been performed in the past on two models of the

penstock for the reversible hydro-electric power plant in Bajina Bašta [1-4].

In order to assess a real strength of workshop-quality welded joints, large, welded pre-

cracked specimens are tested in pure tension. These specimens, known as wide plates

(WP), belong to the group of large-scale fracture-toughness specimens. The effects of

residual stresses, always present in the non-stress relieved welded joints, are similar to

those in welded structures if base metal thickness and welding technology are the same.

Some WP tests were recently conducted in the brittle-to-ductile fracture transition-

temperature range on a single-bevel, K butt-welded joint in a HSLA structural steel [5-7].

Small-scale fracture-mechanics specimens are usually used to assess the apparent frac-

ture toughness of different zones in welded joints. Some of those specimens are stan-

dardised; others are not. The size of standardised specimens depends on welded joint

thickness. Specimens are easy to prepare, the experiments are simple, quick and cheap,

and loadings are low; just opposite to the situation described in both previous points.

Unfortunately, the effect of residual stresses is practically absent from fracture-tough-

ness data due to limited size of the sampled welded joint. Residual stresses are almost

totally removed during welded joint cutting and specimen machining.

For the study of the properties of the heat-affected-zone (HAZ), the welded joint shape

can be specially designed to allow the test of particular HAZ sub-zone. Single-bevel

welded joints are appropriate because the crack tip can be positioned in the interesting

HAZ sub-zone across the weld-joint thickness. The so-called composite notch is used for

fracture-mechanics specimens of double-bevel, X, or V butt-welded joints. So obtained

joint weakest link is presented in [8]. This type of specimen is also suitable to test fracture

properties of simulated microstructure of welded joints.