Can we weld SS304 to P91? 1

[replica]

Dear all,

Can we weld SS304 to T91? What is the filler metal used and do we need PWHT?

Any comment is highly appreciated.

 

[pradipgoswami]

Yes, there're a number of ways, Could you please specify the design conditions.
Thanks.

Pradip Goswami,P.Eng.IWE
Welding & Metallurgical Specialist
Ontario,Canada.
ca.linkedin.com/pub/pradip-goswami/5/985/299

 

[weldstan]

Yes but I prefer to use an intermediate material between the two especially if the service temp is a 1000 F and greater. I have see directly joined using Nickel base filler metals and buttering fail in as little as 5 to ten thermal cycles.

 

[metengr]

Agree with weldstan. You need a transition piece to avoid premature failure between these dissimilar metals.

 

[pradipgoswami]

Hi Replica,
The suggested technique as above involves use of Transition or PUP Piece. This is one of the recommended technique for dissimilar weld joint between Grade-91 # Austenitic Stainless Steels. The practical difficulty could be availability of reliably welded PUP piece from good supplier.
The other recommend option is use of Nickel base welding Electrode, E-NiCrFe2 or E-NiCrFe-3 with the following technique. This is also suggested by EPRI- Ref,1026584,Creep Strength–Enhanced Ferritic (CSEF)Steel Welding Guide.

* P-91 components weld edge preparation(WEP) bevels should be buttered with the above grade electrode or matching filler metals(Typically, 3/16" after machining/grinding is recommended) .
* Perform PWHT of the WEP bevels to Code requirements.
* Perform welding between P91(buttered) to SS 304 with the same grade electrode.
* No further PWHT would be applicable.

A PQR to Sec-IX (QW-283) should be established following the above suggested welding sequences and techniques.

Hope this would help.
Thanks.


Pradip Goswami,P.Eng.IWE
Welding & Metallurgical Specialist
Ontario,Canada.
ca.linkedin.com/pub/pradip-goswami/5/985/299

 

[replica]

Thank you very much for a valuable comments.

The 304 stainless steel spacer is welded to the final superheater boiler tube (T91). 304 is used as a spacer as it can withstand the flue gas temperature of 1200C without cooling. The weldment cracked at the preferred location at HAZ region of base metal(T91) side (tempered HAZ zone) about 400 micrometer from the fusion line. At least five weldment from the same tube had experienced the similar failure mode.

Hardness reading at the base metal (T91) of the cracked locations show the hardness of between 171 to 175 Hv The hardness reading at the base metal of the good weldments are between 188 to 200Hv.

We also replicated the failed locations and found that the microstructure exhibited tempered martensite typical for T91 but with many creep voids. We concluded that the failure was due to creep but the actual reason for the failure to take place at this preferred location (400 microns from fusion line) is unknown. We will carry a proper failure analysis work to identify the root cause of the failure.

pradipgoswami ...thank you very much for you valuable informations. I will look for the welding procedure that had been used for the components. I will also look at the type of filler metal used.

 

[metengr]

Replica
For existing 304 spacers on boiler tubing, I would use a weaker or undermatched filler metal to have the failure on the spacer side of the tube fail first to avoid failure in the tube itself. This is a common design for boiler tube spacers. You do not want cracking in the tube side HAZ if you can avoid it.

The other option was to have Gr 91 tabs welded to the tubing and N&T to provide a platform for joining the 304. This way the DMW is off of the tube pressure boundary.

 

[weldstan]

Agree with metengr. The very low hardness of T91 tubes is disconcerting and premature failure in Creep would be expected. How long has the boiler been in service?

 

[replica]

weldstan ...the tube was in service for 10 years before it failed at the weldment between the spacer and the tube...

metengr...thank you very much for the suggestion...we used ERNiCr-3 as an electrode.