Fatigue analysis and normalized material

[jtseng123]

Dear all,

A spec from an engineering firm asking carbon steel to be normalized for cyclic service. Based on Div 2 Part 5, the fatigue analysis has nothing to do with whether the material is normalized. Also, the algorithm behind FEA to do stress analysis is just stiffness matrix, which only involves "E", the Young's modulus, and "I", moment of inertia, nothing to do with normalized or not.

Yes normalized material is better with a cost, but I don't think it is necessary nor can it increase fatigue life from code or FEA. What is your thought ?

 

[TGS4]

That is very odd - I have never seen that before. I agree that it doesn't seem to make much sense.

 

[metengr]

Typically, normalization heat treatment results in uniform grain texture and mechanical properties (slightly higher strength and toughness (Charpy) with carbon and low alloy steels. So, this may have been where we need a fatigue analysis and by the way the material is supplied as normalized versus annealed or hot finished for service conditions.

 

[jtseng123]

I think any heat treatment on the material before welding has no merit at all. If it has any merit, the FEA and Code have to be re-written. It is only to let someone who does not know how the code is written or how FEA is programmed to feel "good" by asking normalized material.

We all know that the major concern for fatigue analysis is at the discontinuity areas. The best way is to perform PWHT to remove the residual stress due to forming and welding, such that the vessel is stress-free prior to operating loading and the fatigue cycles can be accuratly predicted, rather than asking for normalized materials, which has no advantage to remove residual stress nor providing better theoretical fatigue life from code or FEA.

 

[metengr]

I think any heat treatment on the material before welding has no merit at all

What? What do think are bulk mechanical properties in normalized and quenched and tempered plate, bar, castings remote from weld regions? Why is there subcritical post weld heat treatment by Code versus a mandate to reheat treat after welding (normalization and Q&T)? Better think again.

 

[jtseng123]

metemgr,
You are misunderstading my point. TGS4 has pointed out that askig normalized material for cyclic service is very odd, that is also my past experience that I never see people asking for normalized material for cyclic service until recently I saw one company's spec asking for that.
Simply put, plain carbon steel is good for cyclic service based on Div. 2 and finite element analysis. There is no need to use normalized material to improve bulk material property. If there is anything to do, I will say PWHT to kill residual stress from forming and welding. However, no one will do that because it is not madatory by code for cyclic service. I believe TGS4 will understand my point much well because he is Div 2 expert.

 

[metengr]

Understood. It may be that the request for normalized material is unrelated to fatigue analysis and could be related to fracture or notch toughness requirements in addition to fatigue (cyclic service).