API 5L linepipe Impact Testing

[Kabootarman]

I have also posted this in ASME section, but I was advised that I might have better response in this section.

ASME B31.8-2020 specifies fracture control requirements for NPS 16 and above pipe when hoop stress exceeds 40%.

- Does anyone have experience with ordering API 5L linepipe when this is the case?
- B31.8 refers to Annex G of API 5L for Charpy energy criteria calculations. Annex G mentions a few methods, which one is the most commonly used in practise?
- Do you typically aim to keep the hoop stress below 40% when performing pipe w.t calculations to avoid fracture control requirements or this is typically not an issue when ordering API 5L PSL 2 pipe?

Please share your insights.

 

[akpipelineengr]

1. Yes, Charpy impact tests are common for line pipe.
2. Any of the methods in Annex G or 841.1.2(c) are acceptable. The values given in G.7 are commonly used because they already did the math for you (assuming you meet the limitations in G.7.1).
3. No. You will typically design at the highest design factor you can. Charpy tests are orders of magnitude cheaper than thicker pipelines. You don't indicate your design minimum temperature, but achieving the minimum energy values is no problem for most design temperatures. Temperatures below -20F require some special attention and cannot be achieved by all mills or processes.

 

[Kabootarman]

Thank you for your response, this is helpful.

 

[SJones]

Designing for 40% max hoop stress will kill a large number of pipeline projects. Generally, the Battelle Two Curve analysis is adopted by major operators. If your gas composition will fit the methane limits of DNV-ST-F101 you can get some ‘off the shelf’ Charpy values from that standard.

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.

 

[Kabootarman]

Thank you for your insight. Battelle two curve method is limited for pressures upto 1740 psi. For anything higher, is G.11 (full scale burst testing) the typical approach?

 

[SJones]

No, simply because it is expensive, time consuming, and will rely on the pipe having been manufactured.

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.

 

[Kabootarman]

Thank you, appreciate your input.

 

[SJones]

Battelle two curve method is limited for pressures upto 1740 psi

Try a Google search on "application of battelle two curve model to high pressure gas pipelines" to get a picture of the amount of work undertaken to work around the original calibration limits.

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.