I'm no corrosion specialist, but it would seem to me that from what I understand, it would more or less depend on the chemical composition and structure of base material and filler metal, the welding process and procedure, and the shielding. Maybe if you post more information regarding these variables, more assistance can be provided. In the mean time, here's what I've read about the subject. I hope it helps: The corrosion resistance of welded joints may differ from that of the unwelded base metal, because the weld zone varies in chemical composition, metallurgical structure, and residual stress levels. These differences are brought about by the welding process itself. To optomize the corrosion resistance of welded joints, it is essential that the proper welding process and filler metal be selected. A welded joint and the base metal may corrode uniformly over the entire surface. The welded joint itself may be susceptible to varying degrees of preferential attack. The weld metal may corrode more or less than the base metal as a result in chemical composition or microstructure, or both. In addition, the heat affected zone may be susceptible to corrosion attack in a specific region, as a result of metallurgical reactions during welding. More than one type of corrosion may occur in the same welded joint. Corrosion can occur uniformly on the weld and base metal, exclusively on either, or in the base metal high or low temperature heat affected zone, which are all visible to the naked eye. On the other hand, microscopic attack, such as intergranular corrosion, which occurs preferentially at grain boundaries, usually with slight or negligible attack on the adjacent grains, and pitting corrosion, which is confined to a point or small area, that takes the form of cavities, can occur at any location in a welded joint.
Offhand I would say that chemical differences between the weld metal and the base metal cause the problem, as swnorris said.
You might want to post the question on either the corrosion doctor's web site, or on the nace.org site. A lot of corrosion engineers hang out there and I sure you will get an answer.
Two mechanisms that might be causing your corrosion problem Weld Decay and Galvanic Corrosion. Google these terms for more info.
Here are a few links that describe weld decay http://www.azom.com/details.asp?ArticleID=99
Galvanic corrosion link
More info please! This is not a question that can be answered without details of the base materials, service temperature/pressure, service environment/fluid chemistry, weld joint configuration, filler metals, weld process, etc. Be leery of assuming a particular mechanism is the cause without considering all these factors.
MB, the base material in one inspection of a nozzle connection where weld seperation and localized corrosion was detected was 516 grade 70 , the filler material used was E7018, double bevel joint, there was a PWHT according to ASME Sec. VIII Div. 1, operating temp and pressure- 300f@80psig , steam service, years in service 25. The feeling I get is that possibly the PWHT was not conducted properly on the new construction of this equipment. I don't usually have access to the original PWHT reports and it wouldn`t matter anyway since the reports can easily be altered by the tech to read what ever he wants. Thanks for your interest, Have agood one!
25 years is a long service life for steam service. The component is probably at the end of its normal life, unless repair is feasible. The 25 year duration seems more than sufficient time for even a weak galvanic assisted corrosion of an anodic weld or HAZ to occur and become extensive. I do not think there is any implcation of incorrect PWHT in this context.
Independent even of chemistry, weld metal, or HAZ, can be anodic (electrochemical pole for preferential material dissolution) relative to the base metal because of microstructural differences (coarse grains, aligned crystal structues, micro-concentration gradients). In this context too, the area ratio of base metal to weld/HAZ works increase the anodic/cathodic potential difference between the poles of the galvanic couple. The composition of the steam and the nature of the local steam flow may have prevented, or interfeared with, the formation of a protective scale or film.
Thanks bonniweldor for your input, this drum measured 6` in dia. x 18` in height and had other problems other than the attack at the weld area, approx. 10% fo the shell suffered erosion in the areas of the distibutor nozzle inlets to the extent that it was below the min. thickness required. The engineers rerated the equipment to lower the t-min. to avoid the costly flush patches and weld overlay that would have been mandatory. Also reinforcement rings were installed on the I. D. to meet the new t-min. requirements. The weld repairs were completed and the localized PWHT was applied to the repaired areas and back in service it went! I sure hope those engineers know what their doing. Have a good one!