African Fusion August 2016

to be on the low side of previously published graphitisation temperatures [4] to [8]. See Table 1 for details of the different welded joints.

Sample number

Heat treatment before weld 600 h at 635°C

Joint design

Welding processes

PWHT

1

Double V-groove Double V-groove 2 single V-grooves

GTAW, SMAW GTAW, SMAW SMAW

600 h at 635°C 12 h at 635°C 600 h at 635°C

2

600 h at 635°C

3

None

4

None

None

None

None

Table 1: Welded joints evaluated during this study.

Figure 4 and Figure 5 depict the joint configuration prior towelding on the graphitised basemetal and graphitised HAZ regions respectively. These samples were tested in accordance to ASME IX, that is, two transverse tensile tests (with a rectangular cross section) and four transverse bend tests were machined from eachweld [11]. The rectangular cross section tensile testswere supplementedwith round tensile tests, for reliableobservation of the plastic flow behaviour. Other tests included hot tensile tests (at 427°C), hardness profiles as well as weld metal and HAZ Charpy impact tests at room temperature. Extensive microstructural evaluation was performed using optical and scanning electron microscopy.

Figure 2: HAZ graphite forming along the low temperature HAZ of a welded joint. [9].

650

Spheroidisation

600

Graphitisation

550

500

450

Temperature °C

400

10

100

1000

10000 100000 1000000

Time (hrs)

Figure 4: Full thickness groove preparation for graphitised carbon steel (joint preparation for samples 1 and 2).

Figure 3: Time taken for half of the carbon in a 0.15%C steel to spheriodise or graphitise [10]. Experimental procedures The objective of this studywas to identify possible repair weld- ing techniques that can be used to refurbish aging C-Mn steel pipelines that contain HAZ graphite. The basematerial was a section of pipe (35mm thickness) that was in service for 35 years at an average temperature of 427°C. The chemical composition and mechanical properties conformed to the requirements of SA-515 Gr 65. The nominal chemical composition was 0.22%C and 0.80%Mn. The mate- rial was removed from service during routine maintenance and found to contain a circumferential weld with significant amounts of HAZ graphite. During the current study, several weld configurations were evaluated: • The original material containing the graphitised HAZ. • A newbutt weld that had beenwelded on homogeneously graphitised base metal. • A partial penetration double-grooved weld, which inter- sected the existing graphitised HAZ. All experimental welds were heat treated before, after, or be- fore and after welding at 635°C for 500 hours in an attempt to induce accelerated graphitisation and mimic long term expo- sure to elevated temperature. This temperature was selected

Figure 5: Half thickness groove preparation machined around pre- existing circumferential groove weld (sample 3). Results and discussion The results of the tensile test can be seen in Table 2 (rectan- gular cross section) and in Table 3 (round cross section). In most cases, the tensile test coupon failed in the base metal. Both tensile samples with a rectangular cross section from sample 4 (as received) failed in the HAZ graphite – Figure 6. It should be noted that the tensile strength of this sample was similar to that of other samples that had not failed in the HAZ graphite (Table 2). For the tensile samples with a round cross section, the reduction in areawas usually between 69 and 74% (Table 3). When the fracture surface intersected HAZ graphite, the reduction in area was 44% (Table 3).

17

August 2016

AFRICAN FUSION