A671 - What Class?

[iam42]

Any help or suggestions would be greatly received.

We have a client that has specified the material of construction to be A671 (he has not defined a class). They are simple pipe spools ranging from 30" to 60" diameters.

They have to be suitable down to -50F.

I have suggested using 516 Gr.60. The problem I am having is determining what class they should be.

Are there rules for determining this?

I would like to go with Class 10, no HT, RT or hydro.

Thanks

 

[metengr]

iam42;
Lets start from the beginning;

1. Actual design conditions for the pipe (P, T)
2. Code of Construction for the application, ASME B31.xx

ASTM/ASME A/SA 671 is a specification for electric fusion welded pipe for atmospheric and lower temperature service.

As a minimum, for low temperature service you should consider
CC 70 plain carbon, killed, fine grain A 516/A 516M 70

You will probably need a Class 50 - quench and temper heat treatment to achive suitable impact values (notch toughness) at -50 deg F. Unless you know design conditions, this will determine final selection of Grade and Class for this specification.

 

[iam42]

Hi Metengr,

Thanks for the help.

The code of construction is B31.3

There are various Pressures & Temps, but a typical example is:

P= 250 psig
T= -50F to 221F
Pipe = 36" Std Sch (.375 Wall)

I considered using 516 Gr. 60 in order to be able to use Curve C from Table 323.2.2A. This gives me a min temp of -55F for 3/8" plate without impact testing.

This stated, i'm still not sure how to determine the required class!

I know performing a quench and temper will improve my charpy's, but is it really require?

Thanks

 

[metengr]

You will need to qualify the weld procedure for impacts, if filler metal is used. See B31.3. For this, heat treatment will be necessary.

 

[iam42]

Metengr,

I have a procedure qualified for GTAW and FCAW to -50F.

The pqr was qualified using 516 Gr 60 Plate. I did not perform any heat treat on this plate. It was 1/2" thick.

Why do you say heat treatment would be necessary?

Thanks

 

[metengr]

What was the original HT condition of the plate for this procedure? Was the plate supplied as N&T from the mill?

 

[iam42]

I just double checked the MTR. The plate was supplied 'As-Rolled'.

 

[metengr]

Ok, Thanks for the follow-up information. Well, if you carefully review ASME B31.3, for exemptions to impact testing using curve C, you need to review Table A-1 for what material/heat treatment/Grade is required.

For Curve C, A 671, the only grade for this curve under ASTM A 671 is CC60. For my above post, I was thinking for curve D, CC70 would be required. Heat treatment is referenced in Table A-1 under note 67, which refers you to paragraph 331 in B31.3.

Read 331 carefully to understand any applicable thermal treatment after forming and welding.

Having reviewed your design conditions for service and applicable NDT requirements in B31.3 assuming normal fluid service, I would specify Class 20 as a minimum, versus Class 10. Why? Forming stresses on this diameter of pipe can lead to residual stresses in the pipe material when coupled with welding residual stresses with no PWHT, could result in a shortened fatigue life. As a minimum, go with stress relief after final forming and use volumetric NDT (spot radiography) to assure weld quality.

 

[iam42]

Metengr,

Thank you very much. Your knowledge of this business is second to none.

I hate to drag this post out any longer, but would you have a different opinion of the class if I decided to use normalized plate?

Thanks again.

 

[metengr]

No. You still have the concern over having no stress relief on large diameter fabricated pipe spools and no NDT of the seam weld.

 

[iam42]

Am I correct in thinking that if I were to roll a cylinder for a vessel (ASME XIII) from the same material, I would not have to stress relieve?

Thanks

ps. I would rather do 100% RT than Stress Relieve the pipe.

 

[metengr]

iam42;
Do not mix Section VIII requirements with B31.3. All I was trying to do was warn you of trying to take short cuts. If you believe residual stresses from welding and fabrication (cold bending) will not be an issue for this piping system, fine. The code only provides minimum requirements, it does not substitute for engineering judgment. I would agree that 100% RT would be of greater benefit than stress relief.

 

[iam42]

Metengr,

I really wasn't trying to take a shortcut. I'm trying to get a better understanding of A671 and a way for me to do this is to compare what I know with various codes and ask "why the difference"..... it does confuse me sometimes...

Again, your expertise has proved very valuable. Thanks

 

[stanweld]

iam42,
The classes in A-671 provide the Engineer greater flexibility in choosing the right material based on cost for particular services. You apppear to be basing the material selection based on pressure/temperature/cost concerns only. There are other conditions that may or should be considered as well; such as fluid corrosivity, fluid toxisity, pressure at the -50F design, severe cyclic service, long distance rail transport, placement near populated areas, etc.

Noting the temperature range shown, was -50 F selected because of ambient conditions or Joules-Thompson cooling with depressurization? Other?

Depending on the answers to some of the above conditions, your original selection may be appropriate.

Based on the observation that you intend to spend the least, you may be purchasing from suspect suppliers. I would recommend, as a minimum, Class 13, simply to provide a warm-fuzzy feeling that the weld seam provided was leak free.

 

[iam42]

The project is for an LNG plant where the processes are Methane, propane and LNG.

I was planning on manufacturing the pipe spools in-house; they are relatively short in length (1 diameter long).

They will all be hydro tested and x-rayed 100%.

 

[stanweld]

On the LNG projects constructed by my company, the Engineer specified A-671 CC70, impact tested at -50F with RT of long seams. In the US, NFPA 59A governs, requiring RT of Circ seams and, as I recall, the long seams.

Think about it. Why would you not RT the long seams when they experience the greater stress relative to the circ seams? Which seam has the greater potential for failure?

 

[iam42]

Stanweld,

I am in complete agreement. If your company manufactured the pipe spools, what class of pipe did you manufacture and what heat treatment did you carry out post-welding?

Thanks

 

[stanweld]

I stand corrected re NFPA 59A Long Seam RT requirement; however, if it applies to your work, review 6.6.3.1. Also review 6.2.2.1. I do not believe that the proposed purchase meets those requirements.

We did not manufacture the spools. We installed the spools at the project site. The Engineer/JV Partner purchased the spools to its specifications.