An Evaluation of Factors Significant to Lamellar Tearing

Abstract

The mechanism by which lamellar tearing occurs during the welded fabrication of steel structures is by now well understood. Fortunately, lamellar tearing is generally detected during fabrication and has rarely been held responsible for service failures. Its consequences, however, have been severe in a considerable number of structures, provoking expensive repairs and subsequent damage suits. Realization has come slowly that lamellar tearing is not solely a materials problem; it is, in fact, a design-materials-fabrication problem and can be resolved only by the co-operative efforts of the engineers from all three disciplines.

The investigation reported here was undertaken to determine how fabrication procedures influenced the risk of lamellar tearing in low-alloy high-strength structural steels and what can be done to avoid or mitigate its occurrence.

The results of the investigation may be summarized as follows :

1. Each heat of steel exhibits an inherent sensitivity to lamellar tearing; this can vary considerably from the surface to the mid-thickness of the plate and from one plate or part of a plate to another.

2. Of welding parameters, the welding process was found to exert the most significant influence on lamellar tearing. Welds produced by the GMA process or the submerged-arc process were consistently more resistant to tearing than those made with the SMA or FCA process. The difference apparently resulted from the presence of hydrogen in the electrodes of the later two processes.

3. Except for very high levels, heat input appeared to be an unreliable method of influencing the tendency for tearing. No trend was observed in a five-fold range of heat inputs with the GMA process.

4. Preheating was beneficial to the tearing resistance of GMA welds, but it was especially useful for SMA and FCA welds in counteracting the damaging effects of hydrogen. The use of preheating in actual service structures must be done in a way that does not increase the total contraction of the joint upon final cooling and thus aggravate the tearing strains. In some cases, peening may be helpful in counteracting the contraction strains in the joint.

5. Experiments in which the hydrogen potential in the arc atmosphere was deliberately varied demonstrated clearly the deleterious effect of hydrogen on lamellar tearing resistance. This behavior has been shown for both the GMA and SMA processes and for both carbon steel and low-alloy steel plates.

6. The use of two-layer buttering either in a pregrooved plate or on the plate surface was shown to be highly effective in suppressing tearing. This benefit was associated with the tendency of tearing to initiate within 5 mm (0.20 in.) of the fusion line of the joining weld.

7. Preliminary tests suggest that incipient lamellar tears buried from view do not lower the static strength of the joint seriously unless they have propagated sufficiently to be detectible by UT inspection. The plasticity of the joint is lowered by buried tears before its strength is impaired.