FEA of Pressure Vessel with Combined Loads 1

[kee0419]

I'm new to analyzing pressure vessel using ANSYS and had few questions:

1) The design pressure of the vessel is only 2.9 psig. I calculated MAWP to be 313 psig and maximum external pressure to be 205 psig, both hot & corroded condition. My manager thinks for FEA, I can just apply internal pressure of 2.9 psig but is he correct? Should I apply the MAWP as internal pressure with maximum external pressure? What about new & cold condition, which has higher pressure values?
2) I am combining pressure, temperature, weight and seismic loads all at the same time in my FEA, is this correct way to analyze? Temperature load alone results in very high stresses.

Pressure vessel I'm analyzing mainly consists of 5" OD top head with 3/4" shell and 3" OD bottom head, vertical vessel to be mounted on the wall.

 

[innovation2]

Don't apply internal and external pressure at the same time, that doesn't make sense. I am not an expert but I would assume you would have to consider each of the different types of loading action simultaneously on the vessel. Each of the load types being described in item 2 above. Design for hot/cold, new/corroded and internal/external pressure.

 

[kee0419]

innovation2,

According to ASME VIII-2, Table 5.2, P = Internal and external maximum allowable working pressure, so I assumed I can apply both loads at the same time.

 

[kee0419]

thank you Fawkes for the clarification, I will run the analysis with different load combination.

 

[TGS4]

kee0419 - you say that you are new to analyzing pressure vessels with ANSYS. Are you new to analyzing pressure vessels, or new to doing so with ANSYS?

Your question seems to me to be very elementary - do you understand how pressure vessels work, and more importantly, how they fail?

 

[kee0419]

Dear TGS4,

I'm new to both, as my experience was using FEA for consumer product design but somehow my company thinks my experience is suitable for pressure vessel analysis. I know it requires experienced engineer to design pressure vessels so your input will be greatly appreciate it.

 

[TGS4]

Glad that you are aware of your limitations. In my opinion, it is more important for you to focus on the pressure vessel side than the FEA side for now. For starters, read through ALL of Part 5 of Division 2. Read it with the intent of understanding the failure modes that are discussed, with less emphasis on the mechanics of how to actually implement all of the analyses.

For starters, pressure vessels have four major failure modes: plastic collapse, local failure, buckling, and cyclic failure (of which ratcheting and fatigue are the subsets of cyclic failure). Now, every evaluation that you perform will be in the context of determining the margin against these failure modes.

To your immediate question - you are to use the design pressure in your analyses. Or the design external pressure (the terminology of this has changed over the 2007 edition and 2008 and 209 addenda - make sure that you are using the latest edition. I would even go so far as to recommend that you use the 2010 edition if this is a Division 1 vessel that you are analyzing to Division 2). If you are doing an elastic analysis, follow the load cases in Table 5.3. That dictates how you actually apply the loads for the evaluation of the protection against plastic collapse per paragraph 5.2.2.

Please come back with more questions...

 

[kee0419]

Dear TGS4,

Thank you for the information. I will read the Part 5 of Div. 2. We are to design the pressure vessel using 2007 edition with 2009 addenda.

So you are saying I should use 2.9 psig design pressure for internal pressure NOT calculated MAWP and MAEWP? But the Table 5.2 describes P = Internal and external maximum allowable working pressure.

 

[TGS4]

No, Table 2 does NOT say that.

The 2009 Addenda modified Table 5.2, so that P is defined as the "Internal and external specified design pressure". Please make sure that you are using the latest addenda of Division 2.

 

[COSTASTSAPROUNIS]

when you have the first combination of internal pressure and all other loads you will apply static analysis .
But becarefull for the external pressure you will apply buckling analysis


Costas J. Tsaprounis

 

[corus]

I'm not sure why you're using FEA unless part of the vessel is outside of the limitations of the pressure vessel design code. I'd read and follow the code hand calculation methods. Where part of the vessel is outside of the code, say where a nozzle is too close to the head, then use FEA to assess it.

Tata

 

[kee0419]

corus,

The pressure vessel I am analyzing has nozzles at the heads and near the heads so we decided to use FEA.

TGS4,

I was looking at the 2008a addenda not 2009b addenda and you are correct about the pressure.
I was told to analyze plastic collapse using elastic method so I followed Table 5.3 load cases. According to 5.2.2.4, I should calculate equivalent stresses using eq (5-1) and compare against eqs (5-2) to (5-4). Can I extract equivalent stresses from FEA results and not use eq (5-1)? How about the secondary membrane + bending, PL+Pb+Q requirement for thermal loads?

 

[TGS4]

Can I extract equivalent stresses from FEA results and not use eq (5-1)?

Yes - provided that you follow the methodology described in Annex 5.A.

How about the secondary membrane + bending, PL+Pb+Q requirement for thermal loads?

I take that to mean that you are referring to fulfilling the requirements of 5.5.6? If so, then you also need to follow the methodology in Annex 5.A.

In both cases, ensure that your SCL is "valid", make sure to follow the guidance in 5.A.3.c.

A question for you - have you read and understood 5.2.1.2, 5.2.1.3, and 5.2.1.4? You say that you were "told" to use the elastic method, but have you satisfied yourself that it is applicable?

Have you read this paper, delivered at this year's ASME PVP conference: "A Novel Comparison of Design-by-Analysis Methods" by Stonehouse et al. It talks about how it is possible that an elastic evaluation can completely miss some failure modes. (I don't see PVP2010 papers are available to buy yet from the store.asme.org...).

 

[kee0419]

TGS4,

I have not read the paper you mentioned.

Referring to Table 5.3, design load combination 1,3, and 6 apply to my analysis. Can you confirm if I am using correct allowable stress for each load combination?
1)P+Ps+D: PL+Pb<1.5S
3)P+Ps+D+L+T: PL+Pb+Q<2Sy
6)0.9P+Ps+D+0.7E: PL+Pb<1.5S

 

[TGS4]

Not quite. The allowable general membrane equivalent stress is S. Everything else is defined per 5.2.2.4.e. However, when you define Pl and Pl+Pb, you need to refer to 5.2.2.2.b and 5.2.2.2.c. Pay special attention to 5.2.2.2.b.2.

For all load cases, you need to satisfy equations 2 through 4. All of them. Every time.

Regarding what you've written for load combination 3, that's a completely different assessment, covered under 5.5.6. Remember that you're dealing with a stress RANGE here, not a strict stress quantity. And this stress range limit must be met for every permutation of load combinations, payign particular attention to where loads alternate from positive to negative.