22.08.2015   Comprehensive nuclear materials

Quantification of Residual Stresses and Strains

4.09.2.1.1 Elastic stress

Stresses in welds can be determined via several computational and experimental techniques. Compu­tational methods are generally based on finite element methods’ while experimental techniques include X-ray and neutron diffraction, hole drilling, and sur­face deformation mapping (e. g., slitting). Details of the application of these techniques can be found in several research proceedings55—57 and recent books.58’59

4.09.2.1.2 Plastic strain

The evolution of automated electron backscatter dif­fraction analysis has made the mapping and quantifi­cation of plastic strains in welds accessible via the scanning electron microscope.43’60-64 Strains can be visualized qualitatively via the intragrain misorienta — tion (Figures 11 and 19(c)) of the diffraction pattern or quantitatively (Figure 13) via the average intra­grain misorientation (i. e., the ‘AMIS’ parameter) of many grains and an appropriate calibration curve. Calibration curves from uniaxial tensile samples for several nickel-based alloys are given in Figure 16.

For reference, the measured plastic strain in sev­eral different welds and a heat-affected zone are compared in Table 2. Appreciable plastic strains (2-4%) occur even in unconstrained bead-on-plate welds and a wide range of strains (^2% to almost 30%) may be found, depending on the precise weld geometry, constraint, and welding practice. An example of the effect of welding practice is

image352

Figure 15 An example of the effect of wire cleanliness on weld quality. The top picture shows two GTAW welds made under identical conditions. The right hand weld displayed poor flow and tie-in and was later traced back to impurities (likely drawing lubricant) embedded in the weld wire. The bottom figure shows an SEM image of the wire surface and Auger maps of embedded calcium, carbon, and oxygen contamination.

 

Table 1 Comparison of some physical properties for elements of interest to nuclear power systems

Metal or alloy

Melting point (K)

Thermal conductivity (W mK1)

Coefficient of thermal expansion, ~293-373 K(106K1)

Reference

Fe

1809

80.4

11.8

Ni

1726

74.9

13.3

[44]

Zr

2125

21.1

5.0

given in Figure 17 for a 2 in. thick, Alloy 690 narrow groove weld made with EN82H filler metal via automatic gas tungsten arc welding (A-GTAW) 42,43 If welded with no ‘repairs’ (i. e.,
autogenous remelting of beads to improve bead — to-bead tie-in), it shows ^5.5% plastic strain near the weld root. This plastic strain increases if the beads above the weld are remelted as shown in the

image353

0 1 2 3 4 5 6

AMIS parameter (degrees)

Figure 16 Tensile data used to calibrate the ‘AMIS’ parameter for several austenitic alloys. The ‘GE Data’ are from Angeliu, T. In Tenth International Conference on Environmental Degradation of Materials in Nuclear Power Systems; NACE: Lake Tahoe, NV, 2000.

 

Table 2 Comparison of the experimentally measured ‘AMIS’ parameter and the calculated plastic strain for several nickel-alloy welds and a heat-affected zone

Weld

Range of AMIS measured (degrees)

Approximate plastic strain in weld or HAZ (%)

EN82H, unconstrained bead-on-plate, A-GTAW

0.9-1.3

2-4

A600/EN82H pipe weld, near root, M-GTAW

1.1-1.9

3-8

A690/EN82H narrow groove weld near root, A-GTAW

1.0-3.2

2-13

Best practice

A690/EN82H narrow groove weld, A-GTAW

3.0-6.2

12-28

Abusive weld practice

A600 HAZ (unconstrained, E-182 SMAW)

1.0-1.5

3-6

graph with strains of ~11.5%, 15.0%, and 16.5% with 1,2, and 3 simulated ‘repairs’ above the weld. As expected, high levels of plastic strain lead to increased yield strength, decreased ductility, and increased susceptibility to stress corrosion cracking.