PWHT of duplex stainless steels 6

[springwaterbr]

Hello, all,

I've read around these forums and have come across a seeminlgy consistent opinion that post-weld heat treatment is not a recommended pratice for duplex stainless steels on account of the possible formation of a deleterious phase in the material.

What is making me uneasy about this whole topic is how both ASTM A790 and A928 require welded pipe to be heat treated, unless otherwise explicity ordered.

What am I missing?

Our pipe spec calls for A790 pipe, and a vendor is offering to furnish welded pipe without the heat treatment. Given that the ASTM standards explicitly allow for the purchaser to make a requirement on the necessity of heat treatment, in what situations would it be necessary/recommended?

Thank you.

 

Without delving into those standards, I can say that not all heat treatments are equivalent.
What I can say about duplex SSs is that post-weld stress relief as it is typically understood (e.g., 1100°F) is forbidden for the reasons you state in the first para.
Duplex SSs, welded or not, come with a service temperature limit of around 600°F.

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

 

[EdStainless]

No, no, no. The purchaser may ask for the omition of the heat treatment. The default condition is heat treated. This heat treatment is not PWHT, it is a full anneal.
If it is supplied un-HT the documentation shall show that the plate was properly heat treated, and that the finished pipe meets the A923 requirements.
All duplex pipe and tube should be tested to A923 B or C. All, every time.

Are you welding any of this pipe in your system? Have you tried qualifying welds to A923 when they are done on HT-0 material? My hunch is that you can't.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[springwaterbr]

Thank you for your responses.

@ironic_metallurgist, is that to say that heat treatments performed at a higher temperature than that usually associated with PWHT stress relief are acceptable?

I am still in doubt about this issue:

ASME B31.3 has this to say about requirements for postweld heat treatment of duplex stainless steels: "Postweld heat treatment is neither required nor prohibited, but any heat treatment applied shall be as required in the material specification."

@EdStainless, I've so far understood "postweld heat treatment" to mean any treatment involving heat applied after welding, regardless of duration, intensity, and prescribed cooling process of said treatment. Is it the case that in the industry, in English speaking countries (I'm not from one), that the meaning of the term "PWHT" is usually restricted to a subset of possible heat treatments? In any case, isn't B31.3 defining the PWHT for DSS to be that which is prescribed in the material standard, such as A790 and others? As such, is the full anneal the applicable heat treatment for postwelding? Is my understanding correct?

Also, I've checked that A790 references the requirement that DSS pipe be tested against A923, in the event that heat treatment is omitted, while A928 makes no mention of it; it simply states that such pipe must be appropriately marked, as does A790. What gives?

 

[EdStainless]

I wouldn't touch A928 with a 10' pole. It is really written for pipeline type requirements.
Coming from a SS background I think of PWHT as being sub-critical, in other words not a full anneal. But I am not sure that that is technically correct.
And for virtually all stainless alloys this can pose serious problems.
That is why you have the statement in B31.3 saying any PWHT must comply with what is required in the material spec. I have seen people building spools in a shop and giving them a full anneal after welding. The other issue with duplex SS is that they are very quench rate sensitive, just getting the temp correct does not finish the job.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[springwaterbr]

You mean a sub-critical heat treatment is not advisable for stainless steels in general, including duplex alloys, whereas a full anneal would be (being the only applicable one)?

As to the pipe the vendor is offering to furnish, if it won't be heat treated, it must pass tests per A923, as stated in A790, correct? I'm coming to understand that the PWHT for DSS people here were advising against is not a full anneal, but rather a sub-critical heat treatment. The full anneal is not only recommended, it is mandatory, unless conformance to A923 is proven. This full anneal would also be the only acceptable heat treatment for postwelding of pipe spools -- if heat treatment is called for -- as per B31.3 -- and it is acceptable, as you mentioned having seen happening. Am I following?

On another note, out of curiosity, is there something in the content of A928, or is it something more historical/pertatining to its actual application in the industry, that indicates it to be pipeline-oriented?

Thank you once more for your help.

 

[SnTMan]

An example, Sec VIII, Div 1, UHA-32, PWHT for Hi-Alloy, 10-H:

S31200, S31803, S32550 PWHT 1900 F (min), liquid quench or rapid cool by other means

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[SJones]

Strike the vendor off the approved list and develop a more robust specification to prevent this from happening again. Here is one to give you a little help:

https://www.iogp-jip33.org/product/S-563v18-12.pdf

It might also pay to brush up a little more on metallurgy and fabrication practices.

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/in/drstevejones

All answers are personal opinions only and are in no way connected with any employer.

 

I concur with Steve.
Duplex is not for beginners; in fact it is a class of alloys I would rather avoid because of the extreme level of controls and testing necessary for maintaining full corrosion resistance.

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

 

[springwaterbr]

Thank you all for your various inputs on this matter.

SnTMan, A790 establishes very similar minimum temperatures and rapid cooling for the heat treatment of those grades of DSS. I gather that the PWHT mentioned in B31.3 for DSS is in fact the full anneal required by these material specifications.

SJones, thank you for the document, it will be of much use. Do you have any other specific references on metallurgy and fabrication practices that you deem especially helpful? As to striking the vendor off the client's approved vendor list, I have nowhere near as much authority to do so... Though if it must be, I'll require they include PWHT in their technical proposal. If they can't comply, I'll turn down their offer.

On topic, I'd like to clarify some final things:

- ASTM A790 requires a PWHT for welded pipe or assurance that the pipe complies with the testing requirements of A923;
- Subcritical PWHTs for DSS is not recommended. If a PWHT is necessary, the minimum temperature at which it is applied must be greater than around 950 °C, in order to prevent the formation of deleterious intermetallic phases.

With that in mind, is there something fundamentally different between the seam weldment of a DSS pipe and the weldment bonding two lengths of pipe to form a spool? I believe that in the latter case, it is not very frequent that a full anneal PWHT is executed. However, why is it a necessity in the manufacture of pipe itself? A790 requires the weld be fillerless. Is it a matter of filler vs. fillerless? I can't exactly see why this would be so great a discrepancy so as to rule out PWHT. What am I missing?

 

[SJones]

One thing to consider is that it is far,far easier to give a pipe a full body heat treatment in the furnaces of a manufacturing facility than it is to heat treat a pipe spool in a fabrication shop.

Search on 'practical guidelines for the fabrication of duplex stainless steel.'

Good welding specifications usually detail requirements for processes, consumables, corrosion testing, etc, etc to a level that accounts for the absence of subsequent heat treatment.


Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/in/drstevejones

All answers are personal opinions only and are in no way connected with any employer.

 

Is it a matter of filler vs. fillerless? I can't exactly see why this would be so great a discrepancy so as to rule out PWHT. What am I missing?

It is indeed a matter of filler vs. fillerless, which I don't think has been mentioned so far.

Matching filler metal for 22-5 is overalloyed to the tune of 9% Ni, which accounts for the cooling cycle experienced by weld metal and helps it achieve the appropriate ferrite/austenite balance matching the base metal. A solution anneal heat treatment negates this approach and will leave the weld deficient in ferrite. So if annealing is planned you would either used a compositionally matching filler metal (I doubt one is available) or weld autogenously. In terms of maximizing the properties of a duplex SS weld, autogenous + solution annealing is the optimum approach, although it is expensive, impractical and often impossible.

On that note, beware of heat exchanger tube constructed by autogenous, high-speed electric welding that is not subsequently solution annealed. There is lots of this cheap material on the market, but I would never consider it because the weld deposit corrosion resistance is significantly compromised by this manufacturing route.

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

 

[EdStainless]

IM has hit the nail on the head, autogenous vs filler metal is the key here. For 2205 the welding is done with 2209 filler to preserve the A:F ratio in the welds.

Yes Springwater, you are on the right track.
The lean duplex grades (20-21%Cr no Mo) are easy to work with and very forgiving, 2205 is not bad in thinner sections, the super duplex grades are difficult even in tubing gauges (<0.100") and heavy sections require great expertise.
Duplex alloys in general are great. They have excellent strength, a useful level of Cl SCC resistance, and they are a very efficient use of alloying elements (pitting resistance vs alloy content).
However they do have a limited temp range (0 - 600F), can be tricky to weld (and still get good pitting resistance), and the processing can change the properties.
One example of this last point is that the anneal temp determines the A:F ratio. With austenitic stainless it just has to be annealed enough, and over annealing doesn't cause harm, while in a duplex annealing at the bottom end of the range vs the top can shift the A:F ratio 15% or more.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[springwaterbr]

Hello, everyone!

I'm sorry I didn't get back to this soon enough. Picking up where we left, I'd like to press on a little longer.

The filler having a different composition which obviates heat treatment post-welding makes sense. However, ASTM A928 specifies E2594/ER2594 consumables for welding UNS S32750 (the pipe spec with which I am involved right now). And those consumables have exactly the same composition as UNS S32750 itself! So it is indeed a compositionally matching filler. The consumables listed for other grades of base material are not matching though. Yet A928 makes no distinction between grades and how one should proceed with respect to heat treatment. Recap: A928 has this to say on this topic:

6.3 Heat Treatment:
6.3.1 Unless otherwise stated in the order, heat treatment
shall be performed after welding and in accordance with the
requirements of Table 2.
6.3.2 If the purchaser desires pipe without heat treatment
subsequent to welding, the purchase order shall specify the
following condition:
6.3.2.1 No final heat treatment of pipe fabricated of plate
that has been heat treated as required by Table 2 for the
particular grade. Each pipe supplied under this requirement
shall be stenciled with the suffix “HT-O.”

So no reference to ASTM A923, as A790 makes, as we've said before. Yet heat treated is still the standard condition, matching filler or not. Some thoughts and questions:
[ul]
[li]UNS S32750 pipe is welded with a matching filler when fabricated into spools. Usually though no heat treatment is specified. Yet heat treated is the standard condition for pipe as per ASTM A928, and the purchaser ought to specifically request that pipe not be heat treated and appropriately marked. Is there something fundamentally different between filler-weldments during pipe manufacture and spool fabrication, given that both employ the same filler, to explain the different approaches to heat treatment?[/li]
[li]A790 specifies fillerless welded pipe. Is there something fundamentally different betweent an autogenous weld and a weld made with a compositionally matching filler? Different temperatures, durations and strains, leading to heat-affected zones with different characteristics, maybe? A790 makes a point to ensure that pipe that hasn't been heat treated comply with ASTM A923, which is a step-up from simply requiring special marking.[/li]
[/ul]

Thank you all for your time and attentive input.

 

[SJones]

Check the nickel content of ER2594. The 9 gives a clue.

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/in/drstevejones

All answers are personal opinions only and are in no way connected with any employer.

 

[weldstan]

Heat treatment of the welded tubular in both specifications is predominantly defined to provide best corrosion resistance. If the purchaser decides that no heat treatment be done for whatever reason (price, delivery, knowledge or lack thereof of corrosion failure potential), the purchaser must state so in the purchase order.

There is a difference in welded DSS tubulars with and without filler metal - thickness of tubulars and volume of melted material in the joint which will affect austenite/ferrite structure and balance.

 

[springwaterbr]

For some reason I had it in my mind that I had checked the compositions of E2594/ER2594 against UNS S32750 and seen them to be the same. I don't know how I arrived at that conclusion. That one is on me, thank you for pointing the higher nickel content. Why though does A928 list UNS S32750 as the UNS number for ER2594?

On heat treatment, I'm gathering that it would not be recommended for filler-welded pipe/spools on account of it throwing off the A:F balance. But why then is heat treated the standard condition in A928? The standard specifies filler-welded pipe manufactured from plate, and recommends the over-alloyed-with-nickel consumable. Shouldn't that preclude heat treatment, given that the plate from which the pipe is made must comply with A240 and would already have received heat treatment in accordance with A480 (the same heat treatment required in A790)?

Thank you.

 

In summary:

1) Welded without filler metal >> solution anneal heat treatment (excepting perhaps lean grades in mild environments)
EXAMPLE: Electric welded tube & pipe must be solution annealed.

2) Welded with filler metal >> enhanced Nickel filler metal + no heat treatment of any kind is permitted
EXAMPLE: Grade 2205 Duplex SS (22Cr/5Ni) is welded with E/ER2209 filler metal (22Cr/9Ni) and is not subsequently heat treated.

I must add that choosing the correct filler metal for ferrite balance is only the first of several challenges in welding DSS successfully. As I said, fabricating DSS is not for beginners.


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

 

[springwaterbr]

2) Welded with filler metal >> enhanced Nickel filler metal + no heat treatment of any kind is permitted
EXAMPLE: Grade 2205 Duplex SS (22Cr/5Ni) is welded with E/ER2209 filler metal (22Cr/9Ni) and is not subsequently heat treated.

@ironic_metallurgist, that's the thing... A928 pipe is welded with filler metal, but the standard still calls for heat treatment, unless the purchaser specifically requests that none be carried out. ASTM A815 allows for fittings welded with the addition of filler metal, and indiscriminately requires heat treatment, filler or no filler, necessarily *after* welding:

5.2 All classes of fittings shall be heat treated in accordance
with Section 6.
(...)
6. Heat Treatment
6.1 All fittings shall be heat treated in accordance with the
requirements specified in Table 2.
6.2 All welding shall be done prior to the heat treatment
specified in Table 2.
6.3 All fittings machined directly from forgings or bars (see
5.12), previously heat treated in accordance with the requirements
specified in Table 2, need not be reheat treated.

That doesn't make much sense from what I've learned from this thread... Any thoughts?

 

[EdStainless]

You can anneal duplex that has been welded with over alloyed filler, but often that filler is selected because you can't HT after welding. You just have to know what filler was used in order to make sure that you get things correct when you HT.
Welding and HT duplex isn't that difficult, but it is a lot harder than working with 300 series material. In many ways it is more like working with high strength steels.
You have to watch heat input, interpass temp, cooling rate, gases for weld and purge, and a few other things all at once.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy