ASME A335 P11 Pipework - onset of creep 2

[DSB123]

Hi All,
Just wanted to ask for some opinions regarding the temperature at which creep should be considered for A335 P11 pipework. Any comments?

 

[metengr]

Typically, ASME BPV II stress tables have values italicized where time dependent (creep/fatigue, creep, stress rupture) become dominant over tensile, time independent properties.

So, for SA 335 P11, I would use 950 deg F.

 

[pradipgoswami]

Hi DSB123,

I'm sure you're familiar with this website:

http://www.ommi.co.uk/etd/eccc/advancedcreep/DSheets05ax.pdf.

It has many free downloads on creep data.I've attached the individual data sheet for European grade Steel 13CrMo4-5/5-5,which is the nearest equivalence to SA 335 P11.The allowable stress data at various creep life(No of Hours) are tabulated. The data may not be 100% authenticated to calculate the exact creep life.However it provides a good overview.

In addition to the above information provided by metengr, these data may be of help.

Thanks.

Pradip Goswami,P.Eng.IWE
Welding & Metallurgical Specialist
Ontario,Canada.
ca.linkedin.com/pub/pradip-goswami/5/985/299
All provided answer are personal opinions or personal judgements only. It's not connected with any employers by any means.

 

[DSB123]

Cheers for the replies, however there is a large variation in the temperature at which creep "begins" . metengr says ASME gives 950 F (= 510 Deg C) for P11 whereas the figure given by pradipgoswami shows 450 Deg C (for an "equivalent" material) . Significant variation. Hence my request for opinions. The system I am looking at is at 480 Deg C Design but operates at around 460 Deg C. It has been stated that the temperature is below the creep regime but for me it is on the "cusp" and the design life is 30 years with virtually continuous operation. Any comments?

 

[metengr]

DSB123
Keep in mind that the definition of creep for ASME B&PV design is based on x% deformation at temperature and stress. The stress values determined by ASME Section II Committee are based on the lowest stress to produce either x% deformation at the temperature and stress, or stress rupture or tensile properties. The lowest value governs. So, simply designing at 900 deg F does not mean creep does NOT occur. It simply means that creep will not reach x% deformation for the stress and temperature selected.

Now, what you need to do is to develop a comprehensive in-service based inspection plan to monitor creep at pipe girth welds and high stress locations at wyes and fittings. Don't get hung up on analysis details that try to pin point an exact temperature for creep deformation, that is a fools errand.

 

[DSB123]

Cheers metengr,
I understand what you say but a colleague (who has now left) stated that the temperature of 460 Deg C and the Design Temperature of 480 Deg C meant that the A335 P11 would be operating below the temperature at which creep needs to be considered. In view of the design life being 39 years and the fact that the system stresses are quite high I beleive you cannot ignore the effects of creep and as you say it needs to be monitored.

 

[metengr]

Yes, indeed it should to be monitored in service, especially if the allowable stress values approach maximum or locally exceed maximum values.

 

[DSB123]

Well metengr I cannot get them to accept monitoring is required. The operating temp is 460 Deg C and the stress levels are circa 90% of the Code allowable for sustained and 90% for thermal (I do not like stress levels this high by the way) and the design life is 250000hrs plus.

 

[Ronin609]

"Cannot get them to accept that monitoring is required"

Are there no regulations (OHS) laws that cover operation of high risk plant? API 579 is a well known standard for in-service inspection. Here in Australia, having an in-service inspection plan or policy (and actually doing, and keeping records of inspection) is part of the legal framework of owning and operating "high-risk plant and equipment" which means boilers, vessels and a few other things. We use AS 3788 as a starting point, but API 579 is great. Seems strange for someone to expect to be able to run a high temp vessel without monitoring its condition.

 

[DSB123]

Ronin609,
Yes there are statutory requirements for inspection of pipework, however in this case during the FEED it was stated that creep was not an issue with the material. However as it is just within the creep range at 460 Deg C operating temp and the "life" of the plant is 30 years and the stress levels are high I beleive the Client should be made aware and should have "creep damage" checks noted in his long term inspection plan so it is not forgotten about.

 

[metengr]

DSB123;
It is unfortunate that the Plant operator does not understand risks associated with potential creep or creep/fatigue failures in piping components. In this case, should a problem develop they will have to react to repair the problem which will cost them time and money. Have you developed a spreadsheet to itemize the cost for inspection annualized over the design life and compare with the cost of lengthy outage and safety? The numbers should speak for themselves.

 

[mrfailure]

Physically, the microstructure starts to degrade from pearlitic to carbides in the matrix, which eventually disappear. During this process creep damage may start to form in weld heat affected zones, depending on the speed of degradation (determined by temperature and pressure). This can be a dangerous problem if welds are longitudinal, but you can still get leaks at girth welds. As noted above, you can calculate your degree of risk of failure over time by using published Larson-Miller parameter data and I'm sure software is available as well. Monitoring (often done using replicas) of weld HAZs is an excellent idea!

 

[DBreyer]

Why not make some quick calculations?
You could calculate the expected creep life at your operating temperature.
Since the operating temperature is expected to be lower than the original design temperature the creep strength will be higher and expected life longer. We are talking about logarithmic relationships here so expected life could by >>2x design life.
Also take into consideration that operating pressure might be way lower than design pressure, which would further reduce the actual stresses and increase the ratio between actual stress and creep strength.

For the calculation you could use the Omega method or Larson-Miller Paramater.

Set an inspection date at 2/3 of theoretical life consumption (inspection before that time don't make much sense as creep can only be detected in later stages of life). Depending on your material and the operating conditions this could be well past the intended lifetime of the equipment which would make inspection obsolete.
Additional

Daniel Breyer
Inspection Engineer

http://lnkd.in/dFcQA-w

 

[davefitz]

Inspection is yet the recommended procedure, and its frequency is dependent on which design code is being used. ASME bases its allowable stress on minimum creep values( which may suggest a 100 yr life for average pipe without stress raisers or overheats or longitudinal welds ), while EU codes base their allowable stress on average creep values , which I think leads to a EU mandatory 100,000 hr inspection. In any case, the actual life of the highest stressed pipe section is not what the designer expects or has included in his calcs; there are in fact stress raisers and defects that are often ignored in the design process . There are weld defects, there is differential settlement between major equipment endpoints, there are incorrect hangar settings , there are startup shutdown thermal stresses, etc , which means in real life there are failures . So, inspecting the pipe sections that are likely to have the stress raisers is a recommended practice.

"In this bright future, you can't forget your past..." Bob Marley

 

[DSB123]

Cheers davefitz,
You always provide good responses and I am in-line with your thinking. Just wanted to get opinions.

 

[AWDMIKE]

Not sure if this reply comes too late or is of any help, but API 579-1, Table 4.1 displays a list of materials and their temperature limit used to define the creep range. The 1.25Cr-0.5Mo material (N&T or annealed) shows a value of 427 C (800 F). I'm not sure what the basis is for the numbers in this table, but some of the temperatures in the table are below the temperatures corresponding to italicized values in Table 1A in Section II, Part D.

As a point of reference, a common carbon steel (A-516-70) is shown at 700 F.