P91 PWHT Question

[iam42]

Hi Guys,

I have just had a question put to me from one of our customers and I am not sure how to answer it!

We supplied them some P91 pipe spools where the welds were PWHT'd for 2 hours per our qualified WPS. Our customer has now added attachments to these spools and now needs to PWHT their welds, This is all good but they have asked me if its OK to PWHT the full spool in a furnace.

I have never encountered this before and my first reaction is to say no as it will mean our welds go through a second PWHT cycle which was not qualified by our PQR. My gut feeling is to tell them that their welds need to be locally PWHT'd.

So.... what do you guys think?

Thanks

 

[metengr]

Yes, it is acceptable to PWHT the entire pipe spool in a furnace to control temperature. Your welds were done under your WPS, the attachment welds were not performed by you based on your post, so as far as you need to be concerned about the owner can PWHT their pipe spools.

 

[EdStainless]

I would rather see them do the whole piece. While your weld properties may change a little, it is better than the random 'edge of local PWHT' that they would get otherwise.
And if they do it right the second PWHT should not hurt.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[DekDee]

Just purely to help broaden my knowledge.
Is there anything detrimental in PWHT something twice, three times or more ?
If the PWHT is below the critical transformation range will it affect the metallurgical properties of the weld or base metal ?
Obviously if you wanted to PWHT something twice you should have a qualified PQR but if you did it three times ? - does anyone have experience as to whether it is beneficial or detrimental ?
Cheers,
DD

 

[weldstan]

I have had PWHT performed up to three times without adverse effects as determined with hardness testing and replicative micros.

 

[metengr]

Most of the problems with Grade 91 and PWHT have to do with the chemistry and original heat treatment. Subsequent PWHT on soft Grade 91 can result in further softening because of starting microstructure. However, if the original heat treatment was correct with proper tempering and microstructure there can be multiple PWHT cycles with little to no degradation in properties when multiple PWHT cycles are applied as weldstan suggested.

 

[DekDee]

Thanks guys

 

[iam42]

Thanks once again for your help guys.

 

[pradipgoswami]

Hi Iam 42,

The question is not one PWHT or two PWHT.What matters is the Cumulative Soaking Time at the PWHT temperature. Section-IX requirements state that PQR should be qualified with minm 80% of the production soak time.

Thus if the original PQR for P-91 production welds were qualified with 4 hrs of soak time , one additional PWHT on the production welds would be covered as per as the Code requirements are concerned.
Also the extended PWHT would not be detrimental to the weld properties, provided all materials properties, welding consumable requirements(Ni+Mn, 1.2% max) are checked prior to any PWHT. However watch out for the hardness, it should not go below 190 HB under any circumstances. If so, bad for creep rupture properties.

In absence of a PQR without any added time , you may have to qualify another PQR with 2+ hours of soaking time or else adopt, local PWHT as per AWS D-10.10.Note localized PWHT may cause overheating if the heating bands are properly controlled which may cause softening or over tempering.

Thanks.

Pradip Goswami,P.Eng.IWE
Welding & Metallurgical Specialist
Ontario,Canada.
ca.linkedin.com/pub/pradip-goswami/5/985/299
All provided answer are personal opinions or personal judgements only. It's not connected with any employers by any means.

 

[weldstan]

The 80% criteria only applies when impact testing is required.

 

[Guest102023]

EdStainless said "And if they do it right the second PWHT should not hurt."

Therein lies the problem. Very often fabricators burn up the useful PWHT capacity of P91 in the first shot by (I) not tailoring the hold temperature to the exact composition; and/or (ii) not controlling the temperature closely enough.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[davefitz]

Some other thoughts:
- review published AWS guidelines ( D10.10) on monitoring parent metal directly under the heat treating coils when locally PWHT'g to avoid over-tempering that parent material; apparently it becomes an issue when either bringing the joint to PWHT temperature too quickly or when the coils are spaced farther than 6" from the weld zone. See published articles by Bill Newel.
-There is an apparent correlation between PWHT , time , and weld zone hardness ( via LMP Larson miller parameter) , as per Vogt. See published articles by Bob Anderson in Combined Cycle Journal. If the simple objective is to both meet code rules and retain a weld zone hardness in the range 192-270 Hb following multiple PWHT's then use of this LMP correlation may be helpful .

The P91 high temperature creep strength is related to maintaining its metallurgical crystal structure defined as "tempered martensite with finely dispersed MX precipitates"; the precipitates serve to block the growth of creep microcracks. Over-tempering causes these MX precipitates to agglomerate and then they would no longer be "finely dispersed" leading to a locally weaker zone in the piping.

"In this bright future, you can't forget your past..." Bob Marley

 

[weldstan]

In light of daefitz, metengr, brimstoner and Ed, the potential for PWHT to be incorrectly performed dictates a cautious approach to multiple PWHTs. In many cases, I have tailored PWHT controls beyond those of Code and AWS recommendation to assure acceptable properties. Manufacturers, pipe spool fabricators, contractor/installers, owners and engineers all need to be on the appropriate page in this regard.

 

[racookpe1978]

Also, you MUST properly and adequately support the WHOLE ASSEMBLY during the PWHT heatup, hold, and cooldown times. The metal (at the temperatures we use for PWHT) becomes red hot, and sits for several hours at the high a temperaturs. If the alignment is critical, if an eccentric load (a heavy nozzle or pipe leg) sticks out from the PV to any length at all, or if an internal component is not properly braced, as the whole assembly heats up, it will sag and distort. Machined surfaces (nozzle flanges sometimes or close-toelrance grooves or slots or valve body sealing surfaces) may be distorted, discolored or become misaligned enough to cause leaks.

Schedule your machining after PWHT - which usually happens anyway since many WPS prevent cooling the weld assembly between final welding and PWHT heatup.