SS stub-in to CS pipe in steam service

[scotty36]

On the advice of a Vendor, we have welded a 316SS piece (essentially pipe) directly into a CS header pipe (DN300 Sch 40) in temperatures operating up to 400degC. Using a 309L electrode.

When we do our finite element analysis of the junction, we see yeilding (at operating temp) local to the weld zone due to dissimilar metal expansion.

This result depends greatly on the assumptions made in the analysis. If the properties of the two materials blend across a wide enough area (ie smooth transition), there is no yeilding. However, if we model the juction as only three material properties (316, 309 & CS with no blending) we get a much higher stress.

Does anyone know how much weld dilution is realistic? Is there a reference that I can use for this?

 

[davefitz]

Your finite element model is correct. The proper weld detail to avoid yielding is to use a wide weld pass, using an incolnel weld consumable. To ensuer adequate weld penetration, there will be alot of spatter, so using an automated orbital welder is recommended,as a manual welder will turn down the amps to reduce spatter, but that would prevent adequate penetration.

 

[metengr]

The first issue related to your weld joint is the use of E309L in a dissimilar metal weld (DMW)joint that operates at or above 700 deg F. The weld filler metal should have been Inconel 82 or 182 to avoid failure from carbon migration (local creep damage along the weld fusion zone due to carbon migration into the higher chromium alloy) and to provide a better match in the coefficient of thermal expansion in service between the austenitic material (316 SS) and the ferritic material (carbon steel). Also, have you considered corrosion in service?

The amount of weld metal dilution is based on the welding process. For example, SMAW provides a 30% base metal dilution rate versus GTAW (TIG) that provides about a 40% base metal dilution rate.

The best reference that I am aware of is as follows;

"Pay Attention to Dissimilar Metal Welds - Guidelines for Dissimilar Metal Welding" by Richard Avery. This paper is located at the Nickel Development Institute Web site.

 

[EdStainless]

You don't say what size these parts are, but I'll guess that they are heavy.
I have see people use multiple weld passes in high temp cases to get the blending. You would be better off with a high Ni filler (82) on the CS. You could do this as a local overlay and then weld the 316 to it using 309. Or you could use multiple passes of 82.
Read Avery's paper. There are also some EPRI studies on this, if you have access to them.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm