Vessel PWHT 3

[blynn]

Can anyone give me any info. on how Tensile stress is affected after multiple PWHT's. I have a vessel (SA-516-70), that due to poor QA/QC and carelessness will have to go through a 3rd heat treatment. Any help will be appreciated.

 

[metengr]

blynn;
The tensile stregnth of the Grade 70 steel plate may decrease slightly as a result of multiple PWHT cycles. However, it is not the tensile strength you should be concerned about, you might want to consider impact toughness, if applicable in your application.

Typically, tensile strength for SA 516 Grade 70 plate comes in well above the minimum requirement in SA 516. So, multiple PWHT cycle will not drop you below the minimum ultimate tensile strength requirement of 70 Ksi. Impact toughness, if applicable, is more of a concern and should be evaluated.

 

[vesselguy]

Good question.

To know the exact answer, I would ask the shop to take a sample plate material (read left over pieces) and have a lab do a simulated PWHT at total simulated time to find out EXACTLY what the mechanical properties will be.

I learned long ago, not from a paper but from in-house experten and from the fabricators, that the mechanical properties can not be guarantee for more than 3 PWHT cycles. In the past, we had asked in our spec to ensure the vessel mat properties can be assured for 4 PWHT cycles and the fabricators come back saying the mill cannot guarantee that, and they took exception to the specification. More recently, equipment that were bought from Italy were able to guarantee 4 PWHT cycles. However, I did not look more into detail if they were just blowing wind.

I now do recall seeing a paper long ago that showed tensile strength drop considerably with prolonged heating time below transitional temp; I have to remember to go find it tonight.

blynn,
If your vessel is going through its 3rd shop PWHT, then it probably won't have much time left for the Owner have the requsite ONE future field repair. I suggest you hire a metallurgist to review the case and formally advise your project. Also, fire your crappy shop inspector who been sleeping on the job to let the fabricator get all this shoddy work pass by.

 

[metengr]

blynn;
The web site from Mittal steel contains a brochure that explains the effect of PWHT time and temperature on mechanical properties of SA 516 and SA 387. Mittal has done a fine job of using the Larson Miller parametric equation to show the effect of time at temperature on tensile and yield strength behavior.

Some things to keep in mind when reviewing this data;

1. You must review the original mill test properties of the steel to see how far above you are from the minimum requirements in SA 516.

2. Focus on the graph of Figure 6 on page 4 of the brochure. The % of original tensile is what you need to understand, this relates to the drop in the TS as noted on the mill test report. If for example your original TS is 78 Ksi, you can effectively drop to 70 Ksi - ASME B&PV Code does not give extra credit for strength that exceeds the minimum - 70 Ksi.

3. You must know the aggregate or total time at temperature, not specific the number of cycles!


All may not be lost regarding tensile properties. As I stated in my first response, notch toughness degradation is more of my concern.

http://www.mittalsteel.com/document...tBrochure/mittal pressure vessel brochure.pdf

- Supplemental Result

 

[stanweld]

I agree with metengr regarding plate material properties. Weld metal selection may prove another matter. Deposited E7018 weld metal and other low carbon equivalent weld metals may not meet the minimum specifed tensile properties, especially if the hold temperature approaches 1200 F at the multiple hold times.

 

[SJones]

Don't forget that the WPS needs to have been qualified with at least 80% of the aggregate time at temperature. If that has now been reduced below 80% (accounting for any further PWHT that may, or may not, be applied) - it's a requalification based on the supplementary essential variables of ASME IX (which is the toughness element that metengr is alluding to).

Steve Jones
Materials & Corrosion Engineer

http://www.pdo.co.om/pdo/

 

[blynn]

Thanks everyone. We are having them take a test coupon and do a tensile test. Additionally, I found some information much like that in the Mittal steel brochure in an old Lukens catalog using the Larson Miller Parameter. I'll let everyone know what happens. Thanks again for everyone's input.

 

[vesselfab]

UOP, a process design company, as well as a few others have started putting in their specifications, a clause that materials must have tests run on multiple PWHT cycles. quanties of cycles depend on the anticipated tests. you must have all for anticipated standard code required pwht, on extra for an anticipated shop repair and then another one for an anticipated field repair. Usually 3 is enough. I have seen tensiles and yields start 8K psi above the minimum allowed and wind up below code specified minimum. It's hard enough to meet on plate, but when dealing with common manufactured fittings such as flanges and ells, you have to be careful to match heat numbers as much as possible since you have to by sacrifical parts for testing.

 

[decode]

BLYNN, It'll be useful to know why you are repeting PWHT, there are a lot of ways out to avoid it or just PWHT partially in case of some repairs. In general the Code does not require the whole vessel to be PWHT. Please advise us!

 

[rhg]

If your vessel has "poor QA/QC and carelessness" you must discard this vessel, use it as storage tank and make a new vessel with "good QA/QC"

rhg