Weld repair of Super Duplex casting (ASTM A996 5A/6A) 1

[billbusy]

we have received a pump casing casting from a foundry. it seems they did some weld repair on it.
The foundry did the hydro-test (1000PSI, 15mins) and it passed.
After we received the part and we redo the test after a few days.
It leaks around the weld repair.

What do you guys thinks. why it can hold the pressure in the 1st test? Is it because they did do proper PWHT after the repair?

Thanks

Past: Oil & Gas industry in Canada.
Present: Desalination pump manufacturer in the US

 

The foundry probably attempted to repair an area with shrinkage porosity or other defects and it was not fully successful. Not surprising given the quality of welding foundries are notorious for. I suspect they did not get all of the defects out, and when it got too deep they just started trying to weld over discontinuous metal hoping it would eventually seal.

I would first question the veracity of the hydro test - was it witnessed and is there documentation?
Demand documentation on the repair - how was NDE done, what was found, how was welding done (including filler metal used). All NDE operator and welder qualifications and procedures.
Do your own investigation; i.e., NDE. Excavate if necessary (sounds like it probably will be anyway).



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

 

[billbusy]

Thanks ironic metallurgist.
You are right, I believe it is a porosity area caused by poor casting design.
I believe they did the hydro as they provided the photos and it is 1st order which also make no sense for them to skip ruling out bad parts.

They didn't do NDE after the repair. We are going to send it back either for cutting or NDE.

The question is why it passed in the 1st hydro? Is it because they didn't do proper PWHT(need to verify) or any other possible reasons?
Do you think the Hydro may hurt the part as I heard lots of stories about super-duplex (for example, one part we machined it right and after a few days in warehouse, the dimensions changed, I am still thinking why as I wasn't involved in this specific case, do you have any ideas?)

Thanks

 

[EdStainless]

Did you allow (or prohibit) weld repair?
Was the casting solution annealed after repair?
Was there testing of the microstructure after anneal?

If it leaks it leaks, they didn't repair it.
Probably dirt blocking the leak.

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

 

[metengr]

Hydrotesting should not hurt the part if proper procedure was followed regarding pressure. I would agree with ironic metallurgist. I suspect the initial hydrotest was not performed with the proper hold time after a questionable weld repair. No NDT after weld repair? Was a proper weld procedure used? Any documentation related to size and depth of weld repair?

I would highly recommend you require a repair procedure for review, pre and post NDT, and the hydrotest procedure - pressure and hold time.

 

[arunmrao]

It is surprising that the customer was not informed about salvaging the castings. NDE, salvaging, hydro testing should each have been HOLD points . I am sure there must be a communication gap, and it was Advantage Vendor.

I agree with ironic metallurgist and a star is coming his way.



"Even,if you are a minority of one, truth is the truth."

Mahatma Gandhi.

 

[billbusy]

Thanks Guys.
Yes, there is no proper NDE, no proper weld repair procedure.
it is a production quantity for small pumps from a oversea supplier.
The supplier didn't advise us there is a failure and weld repair.
it is really a wild world. I feel really embarrassed about this order.

Past: Oil & Gas industry in Canada.
Present: Desalination pump manufacturer in the US

 

[EdStainless]

Go back and read the spec carefully (I presume that one was referenced in the order).
In many cases it says that weld repair is not permitted unless approved by the purchaser.
If it doesn't say that then it probably says that repairs need to be done in accordance with ASTM A488.
So you could ask for all of the qualification and testing required by A488.
If they don't have that documentation then the part is non-conforming.
There is also a section on HT after weld repair, ask for that documentation also.
Though in the end the part leaks, reject it. period.

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

 

[metengr]

If there is no material specification or your own specific requirements referenced on the original PO or contract documents, other than a material UNS designation for the casting, you have nothing to stand on.

If you can't find any reference to a material specification, I would walk away from this and start over with a proper purchase specification using a combination of ASTM for material specification requirements and additional requirements for if a repair is required which should require further notification and documentation, etc.

For all I know, this could be for throw away non-critical cast components, and that is fine.

 

[Metalero]

Shape of the indication where it leak? have you inspected with penetrant liquids or magnetic particles? maybe a pic?

 

[bobjustbob]

I am assuming you ordered the castings to ASTM A995. In accordance with A995 weld repair is allowable. Major repairs would require PWHT. Any repair would require a correct WPS, PQR and WPQ in accordance with ASTM A488. If I were ordering it I would prefer to see an ASME Section IX weld program. The only NDE expected would be visual inspection to MSS-SP55 and the reinspection following the weld repair. If you expect anything beyond that it should be spelled out on the purchase order or in your own specification.

If they do not have a proper weld procedure to ASTM A488 and a weld repair was made, the foundry cannot certify the parts to ASTM A995 as they did not meet all of the requirements of the spec. Since it did not pass hydrotesting the part should be returned. If they do not have the correct weld procedure you should not purchase the castings from that foundry.

You should expect to receive exactly what is on the purchase order. In this case the foundry did not provide a sound casting that met ASTM A995. Not because they performed a weld repair as weld repair is valid and expected on castings. When pouring hot metal through air into a sand mold you will have necessary repairs. It is inherent in the process and not necessarily due to poor casting design. If you expect a casting free from weld repairs that should be specified on the purchase order or in your specifications. It will also come at a higher cost and lead time. If you want to be notified of any weld repairs, again it should be specified on the purchase order and will add cost and maybe lead time.

 

Another issue - this is superduplex SS, a very sensitive material. Without even a welding procedure to go by, I would have serious doubts that the foundry has not irreparably impaired the corrosion resistance.

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

 

[pradipgoswami]

billbusy,

Picking up your post quite late. Could you please specify what's Duplex Grade involved? I would say in the first place the only visible defect was repaired, invisible defects were left as it is.

Repair of duplex castings are tricky. Many cast grades does not match the wrought products. Hence choice of electrodes, extent of repair and the requirements of PWHT after repair such as solution anneal are all to be considered.

It would be good to get the oversight of the problem and materials involved.

Thanks.

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