A790 S31803 corrosion test fail per ASTM G48 Method A 2

[ZHA]

Hello All,

I welded a PQR 0.938" thk with ER2209 (TIG). The PQR was to be tested per ASTM G48 Method A (Acceptance criteria - No pitting shall be visible under 20X magnification after 24 hours exposure at 40 C). Weld metal failed in corrosion test(pitting in weld metal). Base metal passed in corrosion test. My max heat input was 58.18 Kj/Inch. Total 13 layers deposited in the PQR. I need to reweld the PQR. What are the precautions to be taken in rewelding of PQR. I am going to get the new weld consumable certified by the manufacturer to pass the corrosion test as per ASTM G48 Method A. What else do you guys suggest.

Regards.

 

[EdStainless]

Have you tested to A923? Even though the two tests look similar they are intended to show very different things.
Passing A923 (either low temp impact or corrosion) shows that you are free from secondary phases.
This is really the test that should be used.
You need to get some clarification, 40C on 2205 welds sounds like a stretch to me. Is this a spec that many people have been able to meet? I sort of doubt it.
I don't have A923 with me but the temp in ferric chloride for 2205 welds is lower than this.

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

 

[ZHA]

We submitted one PQR with A923 but the customer rejected it. He wants to do per ASTM G48 Method A. It is written in the customer specification.

 

[EdStainless]

What is the temp in A923C?
There are some good references published by NiDI on welding and testing of duplex alloys, and papers written by Roger Francis.
I am out of my office so I can't look it up, but this could be difficult.
When you did the G48 you cleaned the weld like you will in production? Is pickling paste allowed? If so then use it. What about passivation, is it to be used, of so then make sure that you passivate your samples after prep. For this test you don't use freshly ground surfaces, in fact the section surfaces should be polished and pitting on them does not count as failure since they will not be exposed in service. I would prep samples and let them sit 3-4 days before testing, trust me it helps. And make sure that you do not go over 40.2C, seriously this is critical half a degree matters in this test. And since this is practice A it is the non-acidified test solution.

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

 

[david339933]

Agree with Ed, 40C is high for 22Cr...should only be 25C....40C for 25CR.

 

[SJones]

I suggest that you renegotiate the test temperature. The client can have G48A, but they shouldn't push for 40 deg C. Tell them to check out ISO 17781.

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/pub/8/83b/b04

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

 

[XL83NL]

You might want to get a copy of API TR 938-C;

https://www.techstreet.com/standards/api-tr-938-c?product_id=1891162

It contains useful guidelines for both design and manufacture (e.g. welding) of duplex stainless steels.

 

[ZHA]

Failed specimen

 

[EdStainless]

Pits in the cross section faces would be failures in A923, but those are not the surfaces being tested in G48.
The cut edges should have been polished to a mirror finish before testing.
Does the customer spec say anything about which surfaces are to be examined?
In G48 the edges are usually ignored.

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

 

[weldstan]

Agree with Ed. The process fluid seeing side, which is assumed to be the root surface would be the surface which would need to pass the test. Or the face surface if that is experiencing the corrosion service.

 

[SJones]

Unfortunately, for oil & gas, the test area has been more harshly set:

5.5.3 Acceptance criteria
All surfaces of the test specimens shall be considered in the test.
The acceptance criteria shall be as follows.
— No pitting at 20 × magnification.
— The weight loss shall be less than 1,0 g/m2 for products in the solution annealed condition.
— The weight loss shall be less than 4,0 g/m2 for welds in the as welded condition.
Penetrant testing is recommended to judge if pitting is present or not.

The key specification issue is the prior pickling of the test pieces. The above criteria are based on the use of pickling. Demanding unpickled leaves the two parties to argue over criteria.

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/pub/8/83b/b04

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

 

[EdStainless]

That criteria would be fine in A923, with all surfaces freshly ground and no pickle/passivate allowed. At the lower A923 temps you would only get attack if you had secondary phases present.
But if they are looking for a corrosion test (that has nothing to do with the service environment) then why would the edges matter? Would you be allowed to polish and pickle the edges?

The test criteria is unreasonable, so it will take a stretch to meet it.
They have taken a test and misapplied it, and because it worked once they think that it is magical.

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

 

[SJones]

Indeed, the specifier has pushed too far in this instance.

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/pub/8/83b/b04

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

 

[pradipgoswami]

40 Deg C is too high for ASTM G48A test for 22Cr Duplex S.S. 30-35 deg C should be the optimal test temperature. The failure is at multiple locations. Hence a switch to Super Duplex Filler would help.

Also the heat input adopted was very high, 58.18 Kj/Inch or (2.3 Kj/mm))is appreciably high for any dupelx S.S. Try to restrict it to 1.2-1.5 Kj/mm max,that's advised by all good engineering practices for welding duplex.It may help.

As the CPT is 40 Deg C it's always advisable to get the filler metal tested at 40 Deg , plus the recorded weight loss within the limits.

Thanks.


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

 

[ZHA]

Thank you all for the valuable suggestions that you guys put in.

Pradip - We used ER2594 and it helped. The PQR passed in the corrosion test. We focused on the heat input and that helped us a lot I reckon.

Thanks Edstainless, SJones, weldstan, XL83NL, david339933 for your suggestions too.

All of you guys have been a great help.