Page 397 - Forensic Structural Engineering Handbook

11.28 MATERIAL-SPECIFIC FORENSIC ANALYSES

TESTING OF MATERIALS AND STRUCTURES

The testing of base metals used in structures was discussed earlier in this chapter, where the
importance of reference to ASTM Specifications both for testing procedures and for the
mechanical properties of products is highlighted. When testing moves from the properties
of steels in structural shapes, plates, forgings, rods, bars, etc., to fabricated structures and
components, such as weldments, bolted connections, subassemblies, and even complete
structures, then there are fewer standard methods of test, and the existing tests may not pro-
vide the needed information.
This is well illustrated in tests for weldability of steels used in structures. Within the
AWS D1.1 Structural Welding Code, welding consumables, process parameters (preheat,
postheat, weld heat input, etc.), and inspection criteria for welded quality for structural
joints are mandated. The AWS D1.1 code also lists approved standard weld joint designs
and procedures by which alternate joint designs may be approved.
Many welded structures rely on AWS “prequalified” welded joints to reduce the quali-
fication tests that need to be run, or use qualification tests that are permitted to cover a range
of plate thicknesses or are made under ideal conditions that do not represent the conditions
under which the welds are actually made. Thus, these tests may not fully ensure that weld-
ment discontinuities such as cracking will not occur. Another problem with these kinds of
tests is that they do not usually separate fabrication weldability, which is an indication of
whether the joint can be fabricated using the proposed procedure free of impermissible dis-
continuities, from service weldability, which is an indication of the performance of the
weldment in service.
There are, in addition, a large number of weldability tests not in any code or specifica-
tion which, although developed for research purposes, are used to evaluate weldments.
Without attempting to describe these tests, we stress that they may measure either or both
characteristics of weldments, and they may or may not represent the service weldments.
Thus to use the results of “weldability” testing in a failure analysis is a complex and often
confusing process.
Another complicating factor in weldment tests is the fact that weldments are a compos-
ite of both microstructures and materials. Weld metals, for a number of reasons, rarely have
the same chemical composition or microstructure as the plates they join, and their compo-
sition and properties are also influenced by the consumables (flux, gas covering) used. In
addition, the heat of welding creates a zone between the unaffected base plate and the weld
metal, called the heat-affected zone, a region heated in some locations close to the melting
point of the base plate and in others, only a little above room temperature. This means that
the microstructure and properties of weldments cover a range that is difficult to sample in
one test. Based on experience, current thinking is that sampling of the base metal, the weld
metal, and the zone in the base metal just adjacent to the weld metal (the coarse-grained
heat-affected zone) provides the relevant information on properties, especially toughness
of the weldment. The heat-affected zone is generally too narrow for tension testing; how-
ever, CVN or fracture toughness testing can be done by careful (sometimes tedious) notch-
ing or precracking in the heat-affected zone.
During forensic analyses it is often necessary to examine failed weldments in much
greater detail than they were initially examined, and tests in the weld metal and the heat-
affected zone may provide critical information relating to fracture characteristics of the
weldment. Sometimes it is necessary to duplicate welds similar to those in the structure in
modified joint designs that have straight fusion lines perpendicular to the plate surface so
that a straight and flat heat-affected zone results. This allows CVN or fracture toughness
specimens to be produced that have heat-affected zones that can be more accurately sam-
pled. Since production weldments do not often have weld joint configurations that meet this
criterion, even these idealized tests may yield data of uncertain value.