Heat Affected Zone FE Modeling

[CA]

Hi! Experts.

I would like to perform a structural static analysis for a T joint made of staneless steel connected by fillet welds.

I am modelling the T joint (fillet welds included) by Tet10 elements.

My question consist in - In order to represent the influence of heat affected zone in the base metal near the fillet welds, I am considering a ficticiuos or reduced mechanical material properties (E reduced) for this regions. In fact, I am considering 10% less in young modulus for HAZ region.

This is approach does it make sense?
Anyone could suggest me any percentual reduced values for young modulus?

Comments, others FE approachs and suggestions are welcome.

Regards,
CA.

Carlos A. Medeiros
Structure Engineer, MSc.

 

[rob768]

Why would you reduce the E-modulus? What you get in the region of the HAZ is, most likely, an increase in hardness and some lower ductility of the mateiral. In addition, you will have shrinkage stress. The modulus of elasticity doesn't change.
You might apply shrinking stress near the weld, but you'd have to figure out (or estimate) how much.

 

[cbrn]

Hi,
100% agree with Rob768. Don't change E, since this property is the one which changes the least (or doesn't change at all) among the various which get modified in the HAZ.
The best way to determine the stress-strain field in the HAZ would be to do a coupled-field thermal+structural transient analysis, if you want to do it by FEM. The constitutive law of the material will be temperature-dependent (shifts in the yield point, for example, or variation of the thermal conductivity), so as you can understand it's really not a simple task (if you don't get exact constitutive laws, for example from a metallurgy expert, all your analysis will be absolutely pointless).
This matter, as far as I know, has not been fully investigated nor understood nowadays. I know that, in the late nineties, the Mech Dept of University of Padova (Italy) was doing systematic researches with the objective to treat the weldments in a similar fashion as the structural discontinuities, but I remember that things were problematic due to the quantity of variables involved (pre-heating, deposition time of the weld material, joint geometry, etc...).
IMHO, you'd better use some general acceptance criteria over the welded joint, as given by ASME, EN-ISO, etc...
In addition, in mechanics most of the time welded components are stress-relieved after welding, for obvious reasons, so you'd better first of all think if this is the case for you, since the complete analysis of the HAZ would become somewhat unuseful.

Regards

 

[GBor]

Agree with the above that you don't reduce the modulus, but you do have a significant reduction in yield strength...often as much as 50%.

Garland E. Borowski, PE
Borowski Engineering & Analytical Services, Inc.
Lower Alabama SolidWorks Users Group
Magnitude The Finite Element Analysis Magazine for the Engineering Community

 

[cbrn]

Hi,
Gbor, I don't completely agree because most of the constructional steels are hardenable and thermic hardening is one of the easiest preocesses to obtain an increase of the yield limit, though at the expense of a loss of ductility and, if "extremized", a brittle transition. Heat altering during welding is, most of the time, nothing else than an uncontrolled thermic hardening. The real problem of the HAZ is the residual stress/strain state, and the real problem of the weldment is the welding efficiency itself. It also depends on the weld type, to be honest...
Or am I missing something?

Thank you, regards

 

[GBor]

Completely agree with you...there are definitely ways of stress relieving the area, but if those steps are not taken, there is a loss of strength. Here's an example from a Welding Handbook:

Typical Mechanical Properties of Gas Shielded Arc Welded Butt Joints:

For 6061-T6 (As Welded, Not Treated, Annealed Properties are used)
Filler Alloy = 4043
Tensile Strength = 27 ksi
Yield Strength = 18 ksi
Filler Alloy = 5356
Tensile Strength = 30 ksi
Yield Strength = 19 ksi

For 6061-T6 (Postweld Heat-Treated and Aged) (For thickness greater than 0.5 inch, 4643 filler is required)
Filler Alloy = 4043
Tensile Strength = 44 ksi
Yield Strength = 40 ksi

 

[cbrn]

Hi,
yes, I understand. It's a class of materials I don't work with. I suppose the same "degradation" occurs whenever the welding heat causes a re-cristallization to an allotropic state which has lower mechanical characteristics.

Regards

 

[rob768]

yield stress and E-modulus are completely different things. Changing the E-modulus changes the stiffness of the material, whereas changing the yields stress influences strength. You can't just interchange them..

 

[cbrn]

Hi,
Rob768, nobody is mixing E and Yieldpoint; since your first reply, we are on the same wavelength.
Hardening, allotropic change, strength degradation, ductility, stress relief, tensile strength... none of these terms copes with E (except, to a certain extent, the allotropic change...).

Regards

 

[CA]

Hi, Guys!

Thanks a lot for all tips and clarification.

In fact, I have no ideia how to deal with this problem.
I only know that there is a loss of strength in the region near the base metal.

I will try to consider all your replays and suggestions.
Mainly, do not change the modullus E.

I hope to see all soon with other theard.

Regards




Carlos A. Medeiros
Structure Engineer, MSc.