Thick-section PQR extension for a pressure vessel seam
A pressure vessel shop is welding a 38 mm thick SA-516 Gr.70 shell in the longitudinal seam. The existing PQR to ISO 15614-1 was made on a 12 mm thick test piece using GTAW root (ER70S-3) and SMAW fill (E7018), single-V, with a steel backing strip, welded in the flat position (PA). It was impact tested at -40 °C and met 27 J average. The production joint is a double-V, no backing, welded in the vertical-up position (PF), using GTAW root and FCAW fill (E71T-1M). The production thickness is 38 mm. The shop wants to use the existing PQR, arguing that the process and filler are 'similar' and that the impact test is conservative because it was at a lower temperature. The client's inspector has rejected the proposal. The material has a specified minimum yield of 260 MPa and the design temperature is -29 °C.
1. Identify the essential variables that make the existing PQR invalid for the production joint.
Read the worked answer
The existing PQR is not valid because several essential variables have changed. The test piece was 12 mm thick; the production thickness is 38 mm. Under ISO 15614-1, a 12 mm test piece covers a limited thickness range around the test thickness, approximately 5–24 mm, so 38 mm is outside the qualification. The joint configuration changed from single-V with a steel backing strip to double-V without backing. Backing and joint type are essential variables because they change the root condition and heat flow. The welding position changed from PA to PF, which is also essential. The fill process changed from SMAW to FCAW, which is a change of process and consumable type, and is essential. The base material group is unchanged, but that does not compensate for the other changes. The shop's argument that the processes are 'similar' is not sufficient; the PQR must represent the production joint. A new PQR is required for the production condition.
2. Explain why the -40 °C impact test on the 12 mm test piece does not cover the 38 mm production joint.
Read the worked answer
The -40 °C test is conservative only in test temperature, not in thickness. Thickness affects restraint, the number of passes, heat input per pass and the cooling rate in the heat-affected zone. A 38 mm joint has higher restraint and a different thermal cycle from a 12 mm test piece, so the heat-affected zone toughness can be lower. The 12 mm test piece was also welded with a backing strip, which changes the root thermal conditions, and the fill process was SMAW rather than FCAW. The FCAW weld metal may have a different toughness from the E7018 deposit. Therefore the 12 mm impact results do not demonstrate that the 38 mm production joint will meet the required toughness at the design temperature. A new impact test on a test piece representative of the production thickness, process and joint configuration is needed.
3. What qualification testing would you require for the production joint, and what are the welder qualification implications?
Read the worked answer
I would require a new PQR on a test piece representative of the production joint: 38 mm thick, double-V, no backing, GTAW root and FCAW fill, welded in the PF position. The test should include transverse tensile, bend, macro examination, hardness survey where required, and Charpy impact testing at the design temperature or the specified test temperature. The impact specimens should be taken from the weld metal and heat-affected zone at the required locations. If the design temperature is -29 °C, the test temperature must follow the applicable code; the shop cannot rely on the old -40 °C result. For welders, ISO 9606-1 qualification on a 12 mm test piece with backing does not cover 38 mm without backing. The welders need qualification on a thickness that covers 38 mm, in the PF position, with a root that demonstrates full penetration without backing. Fillet or partial-penetration qualifications are not acceptable for the full-penetration butt weld.