BHEL Supervisor Trainee study materials
BHEL Supervisor Trainee study resources

P91 Header Weld Rejection and Repair Disposition

BHEL Supervisor Trainee · Put your knowledge to work in context.

All case studies

P91 Header Weld Rejection and Repair Disposition

During fabrication of a P91 main steam header, a 355.6 mm OD × 40 mm wall butt weld is completed with a GTAW root and SMAW fill. The qualified WPS specifies preheat 200-250°C, interpass temperature ≤300°C, and PWHT at 760 ± 10°C for 2 h. Radiography shows a 6 mm linear indication with sharp ends at the root, scattered porosity estimated at 1.5% of the weld area, and a 4 mm lack-of-fusion indication at the sidewall. The project acceptance criteria state: cracks and lack of fusion are unacceptable; isolated, non-aligned linear slag inclusions up to 3 mm are acceptable; porosity up to 2% of the radiograph area is acceptable. The post-PWHT hardness survey gives 320 HB in the weld metal and 285 HB in the HAZ; the project limit is 250 HB maximum. The contractor proposes local grinding of the root indication and re-radiography without re-PWHT.

1. Disposition each radiography indication against the stated criteria. What is the overall weld status?

Read the worked answer

Porosity at 1.5% is below the 2% limit and is acceptable. The 4 mm lack of fusion is unacceptable outright because the criteria reject lack of fusion regardless of length. The 6 mm sharp-ended root indication is not an isolated rounded slag inclusion; its sharp ends and root location make it rejectable as a crack or lack-of-fusion-type defect, and any crack is unacceptable. Overall, the weld is rejected. The hardness results independently make the weld non-compliant: 320 HB in the weld metal and 285 HB in the HAZ exceed the 250 HB maximum, indicating inadequate tempering. The weld cannot be accepted on the basis of acceptable porosity alone.

2. Evaluate the contractor's proposal to grind and re-test without re-PWHT. Include the metallurgical reason and the correct repair sequence.

Read the worked answer

Grinding and re-testing without PWHT is not acceptable for P91. P91 is a 9% Cr martensitic steel; welding and grinding produce a hard, untempered weld metal and HAZ. If the root indication is a crack, grinding may not remove its full depth, and re-radiography may miss a tight root defect. The existing hardness already shows that tempering is inadequate. The correct sequence is: remove the defect completely by grinding or gouging; confirm removal by MT or PT; re-weld with the qualified WPS and preheat; control cooling; perform PWHT at 760 ± 10°C for the specified hold; then repeat hardness survey and RT or UT. Any weld repair after PWHT on P91 requires re-PWHT. The contractor's proposal would leave residual stress and hardness above the limit.

3. What does the hardness result tell you, and what would you check in the PWHT record? State three checks.

Read the worked answer

Hardness above 250 HB means the PWHT did not adequately temper the weld metal and HAZ. Check the actual PWHT chart against the specified 760°C and 2 h hold. Check thermocouple placement and calibration, especially whether the thermocouples represented the hottest and coldest zones of the weld. Check heating and cooling rates and furnace uniformity, and confirm that PWHT was performed after all repairs, not before them. Also verify the hardness-testing method and location. If the records show a valid PWHT but hardness remains high, re-PWHT or replace the weld. Do not accept the weld on hardness alone, and do not allow local grinding without a new PWHT and hardness survey.