Welding Gr.22 to Gr. 91

[iam42]

Hi Guys,

We have a customer who has requested that we weld A182 Gr.22 sockolets to our A335 P91 pipe header.

I will have to qualify a new WPS, but after some "google" research I am not quite sure what the best welding consumable to use will be.

The design temperature for this application is 1065°F

I am finding some older threads that contradict each other on whether to use an ER90S-B3 or an ER90S-B9 wire.

Is there a consensus on the most appropriate consumable to use?

Thanks

 

[viper3982]

We use ER90S-B3 for GTAW, FCAW.
Sub arc we use B9

Just be aware the concerns with dissimilar metals welded to P91; the EPRI group has a LOT to say about concerning temperatures, PWHT, metal chemistry, etc.

 

[Dhurjati Sen]

Dear iam42,

Put a star for viper3982.

As regards to PWHT, you need to follow that of the higher metallurgy, i.e. P91.

Regards.

DHURJATI SEN

https://www.nace.org/people/dhurjatisen

 

This is not a routine procedure qualification.
If you are new to welding P91, I suggest you retain a specialist with direct experience and up-to-date knowledge of the intricacies. You should perform extensive Vickers hardness testing on the coupon.
Mind your heat input and try to effect tempering of previously deposited beads. After welding, perform a hydrogen bakeout soak, then cool to below 200°F (but not room temperature if it can be avoided) and get it into the furnace ASAP.
PWHT holding temperature for P91 is variable depending on the composition; higher Ni+Mn mandates a lower maximum temperature, due to these elements lowering the ceiling for reversion to austenite (this problem has been a focus of EPRI activity). P22 is flexible w.r.t. hold temperature. Minimize restraint on the joint and do not hold higher or longer than necessary.


"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[iam42]

Thanks Ironic,

We have extensive experience welding Gr.91 and so far I have been able to steer our customers away from any dissimilar welds when this alloy is involved.

However, this customer has stuck his heals in and will not budge.

I have read that carbon migration is a possibility when welding these two materials but not being a metallurgist I am not really sure what is going on!

Have you welded these two alloys together and if so what consumable did you use?

Thanks again

 

Carbon migration is the risk addressed by my recommendation to not hold at a higher temperature or longer than necessary.
There is a phenomenon where carbon will diffuse from the low Cr (P22) side of a boundary (the fusion boundary for welds) to the side containing high Cr (B9 weld deposit). I have observed this in the microstructure of T23 tubes welded with B9 filler (which was also excessively cooked). This diffusion happens at and near PWHT temperatures. It can potentially decrease creep strength in P22 material near the fusion boundary.
I would probably go along with the advice to use B9 filler metal, but I see no particular issues with matching the P22 side using B3 filler. Remember that what you generally are trying to match with CSEF steels is creep strength, although the larger concern when joining to austenitic stainless steel is differential thermal expansion. The diffusion problem can be avoided by using Ni-based filler metals, although they are most often used to provide a thermal expansion transition between ferritic steels and austenitic SSs.


"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[weldstan]

I recommend use of the B3 filler metal which will allow for lower PWHT holding temperatures nd thus mitigate carbon migration from the B3 filler into the P91. HAZ hardness of the P91 more readily decreases at temperatures as low s 1325 F while B9 weld metal hardness requires greater temperatures and to respond an decrease to acceptable limits. There was an early, well publicized failure at the decarburized P22 HAZ (near fusion line) on a weldment using the B9 filler and higher PWHT temps required at the time.