Duplex 2205 cap ferrite, interpass temperature and pitting corrosion
2205 duplex stainless steel pipe spools, 16 mm wall, are being welded for a coastal water system. The WPS specifies GTAW root with ER2209 and argon shielding, SMAW fill with E2209-16, heat input 1.0-1.5 kJ/mm and interpass maximum 150 C. During production the welder uses 1.8-2.0 kJ/mm and lets interpass reach 280 C to keep up with the schedule. Root ferrite readings are 30-35 FN, but the cap reads 55-60 FN. The client's corrosion test on a production coupon shows pitting at the weld cap, and impact toughness is marginal. The welder argues that duplex weld metal should have 35-65% ferrite, so the cap reading is acceptable. The project specification requires 35-65% ferrite in root and cap and a corrosion test per the qualified procedure. No nitrogen addition is used in the root purge.
1. Evaluate the welder's argument. How do heat input, interpass temperature and nitrogen affect ferrite balance and corrosion resistance in 2205 welds?
Read the worked answer
The argument is incomplete because ferrite number is only a quality indicator, not a guarantee of corrosion resistance. A cap reading of 55-60 FN sits inside the specified range, but the same weld has exceeded the qualified heat input and interpass temperature and has failed the corrosion test. Those facts matter more than the ferrite number alone.
High interpass temperature and excessive heat input slow cooling and can promote harmful intermetallic phases, especially sigma phase, in duplex stainless steel. Sigma is rich in chromium and molybdenum and severely reduces pitting resistance and toughness. The marginal impact toughness is consistent with that risk.
Nitrogen is also important. Nitrogen stabilises austenite, helps balance ferrite and improves pitting resistance. A root purge without nitrogen can leave the root with a different balance from the cap. The cap may show high ferrite because it cooled quickly, but that does not prove the weld is sound; it may simply reflect a different thermal cycle from the root.
The welder is therefore comparing one number with a range and ignoring the procedure limits and the corrosion failure. The correct conclusion is that the production welding is outside the qualified window and the microstructure is suspect.
2. What immediate actions and revised controls would you require to correct the cap and corrosion result?
Read the worked answer
Immediate actions
Stop production on the affected spools and quarantine them. Review the weld logs and identify which spools exceeded the qualified heat input or interpass temperature. Do not accept them on the basis of ferrite readings alone.
For the affected production welds, the practical remedy is usually excavation and re-welding rather than local heat treatment, because sigma can be distributed through the weld and heat-affected zone. Any repair must use a qualified repair procedure.
Revised controls
Re-establish the qualified heat input range and measure interpass temperature with a contact thermometer before each pass. Set a hard limit of 150 C or the qualified value, with a hold point if it is reached. Control travel speed and pass sequence, and do not let schedule pressure override the WPS.
Use a nitrogen-containing purge for the root, or the qualified shielding regime. Check purge oxygen and maintain the purge until the root is completed.
Require a new production coupon with the corrected parameters and repeat ferrite, corrosion and impact testing. Only release production after the coupon meets the specification.
3. How would you verify the corrected procedure without relying only on ferrite numbers?
Read the worked answer
Ferrite numbers can be misleading because they respond to composition and cooling rate and do not directly measure harmful phases. I would verify the corrected procedure with a combination of tests:
- Corrosion test on a weld coupon, using the method specified in the project specification. The test should sample the root, fill and cap, not just the cap.
- Metallography to check for sigma, nitride precipitation and excessive ferrite/austenite banding. This is the direct check that ferrite numbers cannot provide.
- Impact toughness at the specified temperature, especially if the service is at low temperature or the client requires it.
- Ferrite mapping across the weld, used as a trend indicator rather than a pass/fail criterion on its own.
- Procedure qualification records showing heat input, interpass, purge gas and consumable certificates.
The verification should be tied to the actual production thermal cycle. A coupon welded with a lower heat input than production proves little if the shop returns to the same fast schedule.