TMS Outstanding Student| CONTENTS |
|---|
|
|
High temperature (565°C, 1050°F) fracture toughness tests were performed on welded specimens of 1Cr-1Mo-1/4 V steel with different levels of mismatch between the base metal and the weld metal and the cracks lying along the fusion line. A wide range of fracture toughness values were obtained for weldments, as opposed to a unique value of JIC and a unique J-R curve typically obtained for homogeneous materials. Detailed observations of the crack path within the weldments were made to understand the wide scatter in the fracture toughness behavior. The yield strength mismatch between the base metal and the weld metal was found to directly influence the stable crack path, and hence the fracture toughness behavior. The denomination of "apparent fracture toughness" was used to describe the variability of the fracture toughness in the weld region due to microstructure and mechanical property gradients. The apparent fracture toughness exhibited a minima at a fixed distance from the fusion line for a specific weld. The relative position of the fatigue precrack with respect to the fusion line and the region of low fracture toughness was also shown to influence the measured fracture toughness behavior of the specimen. A frame-work is provided for representing the weld fracture toughness behavior and the associated variability due to microstructural gradients.
The objective of this research was i) to conduct fracture tests on actual weldments of 1Cr-1Mo-0.25 V steel containing a definite heat-affected zone (HAZ) region with microstructural gradients using undermatched and overmatched weld metals, ii) to characterize in detail the stable crack growth path at various locations along the thickness of the specimen and iii) to interpret the fracture toughness behavior of these weldments using the theoretical results described in the studies mentioned in the previous paragraph. Further, to promote extensive ductile crack growth, all fracture toughness testing was conducted at 565°C (1050°F).
Cylindrical tensile test specimens were machined from the base metal and the weld metal regions of each of the two weldments. Tensile specimens were either all base metal or weld metal, with a 25.4 mm (1 in) gage length and a 6.35 mm (0.25 in) diameter. Compact type (CT) specimens were used for the fracture toughness tests. The specimens were machined with standard 1T dimensions (thickness, B=25.4 mm; width, W=50.8 mm). The location of the specimens was either entirely in the base metal or with the crack starter notch located along the fusion line, Figure 1. The specimens were fatigue precracked to obtain sharp initial cracks of crack size to width ratio a/W=0.6. Following precracking, side-grooves of 10% of the thickness were machined along the crack plane on each side of the specimens. The purpose of the side-grooves was to promote an in-plane crack growth and to maintain plane strain conditions over the entire crack front.
All tensile tests were performed at 565°C (1050°F) according to ASTM E8.4 The fracture toughness tests on CT specimens were performed at 565°C (1050°F) using the unloading compliance technique for measuring crack size.5 The JIC and J-Resistance curves (J-R curves) were determined according to ASTM E8135 and ASTM E1152.6 After completion of the tests, some specimens were quenched in N2 and then cracked open. Two specimens of each condition were sectioned along the mid-thickness prior to quenching. One half of the specimen was used for characterization of the crack path. This half of each specimen was cut at three locations across the thickness to obtain crack profiles, allowing the observation of the stable crack path, in particular the location of the fatigue precrack, the angle of growth and the location of the end of the stable crack along the thickness of the specimen. The second half of the specimen was quenched in liquid N2 and fractured to reveal the crack surface.
Microhardness measurements were performed on "as received" T2 material to characterize the variation in properties in the weld region. The microhardness was measured across a line normal to the fusion line, starting at about 3 mm (0.118 in) into the weld and ending at about 6 mm (0.236 in) into the base metal. Three different locations in the weld were tested, each test consisting of four sets of measurements.
Microhardness profiles of the "as received" T2 condition are shown in Fig. 3. The hardness increases significantly in the heat affected zone and in the weld metal next to the fusion line, with the peak at about 200 µm (0.00793 in) from the fusion line into the heat affected zone. This corresponds to the region of material that had experienced recrystallization during the welding process. The width of the heat affected zone obtained in this sample from the microhardness results agrees quite well with that from the microstructural analysis. Thus, any annealing beyond the recrystallized zone must have been restricted to a narrow zone.
 |
(1) |
where 
oBM and oWM are the yield strength of the base metal and weld metal, respectively. A positive value indicates an overmatched weld metal and a negative value corresponds to an undermatched weld metal. Thus, T1 was the undermatched condition and T2 was the overmatched condition. Also, note that the mismatch level even when calculated with respect to ultimate tensile strength is comparable to the one calculated using the yield strength.
| Table I. Tensile tests results for the base metal and each weld, at 565°C (1050°F) | ||||
| Material | Base Metal | T1 | T2 | |
|---|---|---|---|---|
| Yield Strength, o MPa (ksi) | 383.1 (55.6) |
308.7 (44.8) |
493.3 (71.6) |
|
| Mismatch % | - | -19 | +29 | |
| Tensile Strength, uts MPa (ksi) | 414.8 (60.2) |
347.3 (50.4) |
540.2 (78.4) |
|
| Young's Modulus, E GPa (Msi) | 166.7 (24.2) |
168.8 (24.5) |
166.7 (24.2) |
|
Influence of Mismatch on Stable Crack Path
Due to concentration of strain in the weaker material, mismatch in strength can be quite significant in determining the crack path in welds. We will describe the path of stable crack growth for the undermatched and the overmatched weldments separately.
As mentioned earlier, Asaro et al.2,3 had predicted that the propagation of a ductile crack in a bimaterial with a transitional layer will tend to follow the interface between the weaker material and the transitional layer. In this instance, this interface is the fusion line for the undermatched weld metal if one considers the heat affected zone to be the transitional layer. Figure 5 shows the crack path for two specimens from the T1 condition. The end of the stable crack path was located at an average distance of 2 mm (0.079 in) from the fusion line which corresponds to the interface between the heat affected zone and the base metal where the temperature range during the welding process is sufficiently high to cause stress annealing.
The stable crack in specimens with overmatched weld metal, condition T2, grew immediately out of the plane of the fatigue precrack, deep into the base metal, Figure 6. The average distance from the fusion line to the end of the ductile crack was approximately 3.5 mm (0.138 in). In this case, the interface between the transitional layer and the weak metal lay at the interface between the heat affected zone and the base metal, which was measured at about 1.85 mm (0.73 in) from the fusion line. Thus, the crack jumped even beyond the interface between the heat affected zone and the base metal. Microstructural observations of the base metal in the region of the end of the stable crack did not show any particular features explaining why the crack grew in that region. Similarly, the microhardness measurements did not show any significant variation beyond the heat affected zone of the base metal for condition T2, Figure 4. However, the precise location of this interface was not clearly delineated in these weldments. The Asaro et al. analysis was based on strain concentration and did not consider inhomogeneous fracture properties which are very much the case in these weldments. In addition, when the crack turns, mode-mixity begins to play a role which is also not accounted for in the analysis of Asaro et al. Therefore, number of factors can contribute to the deviation of the precise crack path from the predicted trend. It appears, though, that this simple theory does predict the observed trends correctly.
From these observations, one can deduce that the sign of the mismatch determines if initially the crack will grow in the region of the heat affected zone and base metal or along the fusion line. However, it also appears that the stable crack growth tends to move towards the interface between the base metal and the heat affected zone. This strongly points to the presence of a region of low fracture toughness located approximately at the interface between the base metal and the heat affected zone.
Location of Fatigue Precrack
The other main factor influencing the ductile crack path was the location of the end of the fatigue precrack relative to the fusion line. If the end of the fatigue precrack is located close to the plane of stable crack (the region of low fracture toughness), the extent of out-of-plane growth necessary to reach this region will be shorter. This is illustrated by the two specimens from the T2 condition, Figure 6. Since additional deformation energy is necessary for the ductile crack to grow out-of-plane, it can be safe to assume that the location of the fatigue precrack with respect to the region of low fracture toughness will directly influence the fracture toughness of the specimen. In other words, the larger the extent of out-of-plane growth, the higher the position of the J-R curve, and consequently the larger the value of the fracture toughness. This was confirmed by the two specimens of condition T2 shown on Figure 6. Specimen T2-1, with the fatigue precrack located close to the fusion line, had a fracture toughness value of 221 kPa*m (1259 in*lb/in2) and the value for specimen T2-2 was 77 kPa*m (439 in*lb/in2).
The above arguments imply that the measured fracture toughness of the material in the heat affected zone varies with the position of the fatigue precrack. Since this implies out-of-plane crack growth, the characterization of the fracture toughness using J-integral is suspect. Thus, it is more appropriate to use the denomination of "apparent fracture toughness" rather than the absolute value of the fracture toughness which implies a material constant independent of the location of the precrack. To enable a comparison of apparent fracture toughness values between weldments with different width of the heat affected zone, one has to define a location-defining parameter that is equivalent for any weldment. The relative distance ratio r defined by equation (2) allows this type of comparison.
 |
(2) |
An illustration of the calculation of r is shown in Figure 7. The determination of r for a specimen was averaged from several measurements across the thickness of the specimen. Indeed, the non-planarity of the fusion surface can cause a large variation of the values over the specimen thickness. The issue of the non-planarity of the fusion surface will be discussed in more detail later. r values close to one correspond to the region of lower apparent fracture toughness, positive r values close to zero represent the zone near the fusion line in the heat affected zone, and negative r values are in the weld metal. The value of r of all tested specimens is indicated in Table 2, along with their respective apparent fracture toughness values. Figure 8 shows the representation of the apparent fracture toughness versus r, as well as schematic variations of the apparent fracture toughness for both the undermatched and the overmatched weldments. For |r|>>1, the fracture
| Table II. Apparent fracture toughness values and relative distance ratio r of all tested specimens. | |||
| Specimen | Apparent Fracture Toughness [kPa*m] [in*lb/in2] |
r | |
|---|---|---|---|
| Base Metal | 108 | 619 | - |
| T1-1 | 127 | 725 | 0.26 |
| T1-2 | 130 | 744 | 0.26 |
| T1-3 | 448 | 2566 | -0.86 |
| T2-1 | 221 | 1259 | 0.09 |
| T2-2 | 77 | 439 | 0.42 |
| T2-3 | 173 | 985 | 0.41 |
toughness values approach those of the base metal and the weld metal. The weld metal fracture toughness has purposely been chosen to have different values for overmatched and undermatched condition. In between those limits, the apparent fracture toughness of the fusion zone is believed to vary and obtain a minimum value at r=1. For the overmatched weld, the lower bound fracture toughness does not necessarily occur in the base metal. This representation clearly demonstrates the difficulty and danger in comparing the fracture toughness behavior of different weldments when only a few data points are available.
| FIGURES |
|---|
|
Figure 1: Location and orientation of the compact type (CT) specimens.
Figure 2: Microstructure of the weld regions Figure 2a T1 and Figure 2b T2. W denotes the weld metal, F the fusion line, H the heat affected zone, and B the base metal. Figure 3: Microhardness profiles of the "as received" T2 material condition. "A" represents the small region where annealing takes place adjacent to the boundary between the heat affected zone and the base metal leading to softening. (1 in. = 25.4 mm) Figure 4: J-R curves for Figure 4a base metal, Figure 4b T1 weld, Figure 4c T2 weld, and Figure 4d comparison of the J-R curve ranges for the different conditions. Figure 5: Crack profiles of two specimens from condition T1, Figure 5a T1-1 and Figure 5b T1-2. The stable crack first followed the fusion line, then grew out-of-plane to stabilize at a distance of about 2mm (0.079 in) from the fusion line. This distance is shorter on micrograph Figure 5a, but the average over the thickness for that specimen was approximately 2mm. Figure 6: Crack profiles from two specimens of condition T2, Figure 6a T2-1 and Figure 6b T2-2. The extent of out-of-plane growth is directly related to the offset from the fusion line to the fatigue precrack. Figure 7: Determination of the relative distance ratio: r=D1/D2. D1 represents the distance fro the fusion line to the fatigue precrack, and D2 is the distance from the fusion line to the end of the stable crack. Figure 8: Apparent fracture toughness versus relative distance ratio r for both the undermatched and the overmatched weldments. The weld metal fracture toughness has purposely been chosen to have different values for overmatched and undermatched condition. For the overmatched weld, the lower bound fracture toughness does not necessarily occur in the base metal. Figure 9: Crack surface of specimen T1-2. P denotes the precrack, D the ductile crack, and B the post-test brittle fracture. |
| Search | Student Member Application | Other Best Papers | Student Members' Page | TMS OnLine |
|---|
