Sensitization of 316L relative to weld heat input.

[LesT]

I would like to obtain some kind of comparitive data indicating the relationship of sensitization of 316L with various amount of weld heat input. Does anybody know where I can obtain such info?

 

[weesco]

May you check AWS D1.6 and AWS HANDBOOK

 

[mcguire]

Weld sensitization depends on how long the steel is kept in the 1050F to 1350F range. To extent heat input influences that, it influences sensitization.
The low carbon in 316L lengthens the time you can stay in that range to a degree that it is normally assumed to be immune to sensitization. It takes 0.1 to 1 hour in that temperature range to start sensitization.So unless something prohibits you from cooling through that range in that time span, you really have no problem.

 

[LesT]

Thanks for the input.
Due to the critical application of the work piece, the client insists on sufficient proof that no sensitization will occur. In this case one way to satisfy the client will be to make a test piece simulating the weld configuration, cut it up and test for sensitization, or do some thermal modelling of the weld joint.

Due to the geometry of the workpiece, forced cooling or temperature measurements are not practical.

My first conservative choice of heat input would be 1kJ/mm for a corner joint of 6mm plate. I am confident that these parameters will be OK if plate is allowed to cool down between weld passes. The question now is, would 2kJ/mm or 3 or 5kJ/mm also be OK? At what heat input will it start to become risky. Higher heat input will benefit process time.

What I am after is some info of somebody that has done the following test ie.
Welding of a standard piece of 316L plate
with all geometrical configurations and ambient conditions the same, using increased amounts of heat input. Then evaluate the amount of sensitization at each heat input level. The variables in this may be to many to compare with but may help in convincing the client.