If a C-Mn component or structure is normalised after welding, weld metal strength will fall appreciably, unless the weld was deposited with a weld metal of sufficient alloy content that only a modest reduction in strength occurs.
As a weld is essentially a casting, part of its strength, in the as-welded condition, is derived from the as-deposited microstructure. During normalising, which involves heating to around 900°C, the as-deposited structure is transformed to austenite (the high temperature form of ferrite), with a coarser grain size than that of the original microstructure. Consequently, when the structure cools (at a rate which is much slower than it experienced as it was deposited) and transforms to one of the lower temperature forms of ferrite, the weld metal has much lower strength than it had originally. To provide some resistance to the loss of strength, a weld metal should be chosen that contains Mo (or Ni and Mo), and possibly also V. An example of a situation where the issue is likely to arise is during the manufacture of the dished end of a pressure vessel, where normalising may be carried out after the petal ends have been welded together. See reference below. Suitable consumables are available from several welding consumable manufacturers.
If an unsuitable weld metal was used, and the fabrication has been normalised, the only solution is to cut out all the weld metal, reweld, and then subject the fabrication to a post-weld stress-relieving treatment.
Reference
Quinn T A, Garland J G, and Weyland F: 'The development of welding consumables for heat-treated weld metal'; Proc. Conf. on Trends in Steels and Consumables for Welding, The Welding Institute, London, November 1998, 337-350.
See further information about Materials and Corrosion Management.