Reliability of Weld Microstructure and Property Calculations

H. K. D. H. Bhadeshia

Abstract

We depend in our everyday life on the performance of vast quantities of steel, which we use without giving it a second thought. This is possible because the material is reliable and cheap (weight for weight, some 1000 times cheaper than potato crisps). And yet, there are remarkable technologies and sciences working behind the scenes to create ever better steels which can be assembled into awe inspiring structures. Examples include the giant oil--rigs which pepper the North Sea, the oil and gas pipelines which traverse the frozen wastes, and the 101 Tower in Taiwan, which is now the tallest building in the world. These are all made from steel and rely on welding for their assembly.

A weld is a heterogeneity introduced into a carefully manufactured steel. It is a defect which has to be managed. One way of doing this is through a deep understanding of metallurgy, thereby avoiding the engineering disasters of the kind that plagued for example, the Liberty Ships. A weld consists of distinct zones, each of which is the consequence of a particular interaction between heat flow and the phase transformation characteristics of the weld metal and the parent steel.

Out of all these zones, the weld metal is particularly challenging to design because there is little that can be done once the weld is completed. This contrasts with wrought steel which can be processed and manicured using all the facilities available in a modern steel plant. There is therefore, a major industry devoted to the design and manufacture of welding alloys. Academic activity has supported this industry, both in terms of the underlying science and in the creation of quantitative methods for alloy design. Given all this effort, I shall use this lecture to explore whether it is any longer necessary to conduct experiments when developing welding consumables. Are the models sufficiently robust to be exploited by industry without supervision?

The focus of this paper is on ferritic steels which form the bulk of the billion or so tonnes of steels consumed annually. The development of welding consumables involves {\it all weld--metal} tests in which a joint is deposited such that samples can be obtained without dilution with the base plates. It is fortunate that the literature is rich in data from tests of this kind, data which have liberally been exploited in the development of models.

Welding Journal, Research Supplement Vol. 83, 2004, pp. 237s-243s.

Download PDF file of paper.

Download PDF file of Russian translation of paper. Computer Technologies in Joining of Materials, ed. W. Sudnik, Tula University Publishing House, 2005.

Download high resolution figures.


Superalloys Titanium Bainite Martensite Widmanstätten ferrite
Cast iron Welding Allotriomorphic ferrite Movies Slides
Neural Networks Creep Mechanicallly Alloyed Theses Retained Austenite

PT Group Home Materials Algorithms Any Valid CSS!