ASME Section VIII Div 2 - Fatigue Screening

[BradyJr]

Hello,

I normally work with ASME Section VIII Div 1 but was asked to perform the ASME Section VIII Division 2 -Fatigue Screening (5.5.2.3) on a pressure vessel and I'm trying to sure up my understanding of the screening. Hopefully someone on here has some experience and wouldn't mind sharing their knowledge.

1. Step 2 asks to determine the expected (design) number of full-range pressure cycles. What exactly qualifies as a 'full-range pressure cycle'? Is it when the pressure reaches the design pressure, how many times if returns to be pressurized, or just a cycle. My case in particular has two cycles which are performed one after another. Cycle 1 is a 9.9 kg/cm^2 and cycle 2 is at 10.5 kg/cm^2 with no fluctuations during the cycles. The design pressure is 19.5 kg/cm^2. My current interpretation that a 'full-range pressure cycle' would only occur at startup and shutdown.

2. Step 4 asks to determine the effective number changes in metal difference between any two adjacent parts. The pressure vessel is made of all group 1 carbon steel and the process fluid has access to of the vessel components from the inside so I do not anticipate much of a temperature gradient among components (FEA shows well below 28 C differential). My current interpretaion of this is that NΔTE would be zero?

Any insight or suggestions would be much appreciated.

Thanks

 

[TGS4]

1. Correct - full range pressure cycles are start-up/shut-down. Your other cycles would be partial pressure cycles.

2. Provided that you are assured that, during start-up or shut-down, your process inlet temperatures are barely above the metal temperature, then yes.

 

[XL83NL]

Trevor, one more question, if I may. Regarding full-range pressure cycles ...

Say I design some pressure vessel for 0-50 bar(g) design pressure. During operation, Im cycling from 0 to 45 bar(g) back to 0 bar(g), every 5 minutes, for 7 days/week. Consider 2 cases:
- my vessel's MAWP is much higher than design pressure, say 248 bar(g).
- my vessel's MAWP is just above the design pressure, say 62.8 bar(g).

Is this cycles happening every 5 minutes not considered a full range pressure cycle? If not, why only for start up and shutdown?

 

[DSB123]

XL83NL,
Your cycle 0 to 45 to 0 is a full pressure cycle!!!

 

[XL83NL]

@ DSB: why? It's not a start-up or shutdown range. Is it so because it's based on my design pressure of 0-50 bar(g)?

 

[DSB123]

You are going from zero pressure to full operating pressure back to zero pressure. So that is a full pressure cycle!!! Your temperature may not be doing full temperature cycles but your pressure is!!!

 

[pdiculous963]

XL83NL - The way that it is written in Div. 2 is based on design pressure, so I believe technically you would have to consider those full pressure cycles for both MAWPs. In the post-construction world, however, you could make an argument that your Case 1 should not be counted as a cycle (assuming integral construction) as the range is less than 20% of the MAWP. In the end, full pressure cycles and operating pressure cycles above the 15/20% criteria are counted equally in determining if the screening criteria are met, so there is really little difference between a full pressure cycle and an operating pressure cycle (in a screening assessment).

 

[TGS4]

XL83NL - that's actually a very good question. Since the original question was in the context of 5.5.2.3 (Screening Method A), I'll frame my reply in that context.

If your design pressure were 50 bar(g) and your pressure fluctuating from 0-45 bar(g), then my engineering judgement would call that a full-range pressure cycle, per Step 2. However, if your design pressure were 248 bar(g), then my engineering judgement would not call that a full-range pressure cycle. And that 45 bar(g) delta P in the later case would not even constitute an operating pressure cycle, because the range of pressure variation is less than 20% of the design pressure.

{b]However/[/b], I would note 5.5.2.1(c) which states that if the specified number of cycles is greater than 10^6, then the screening criteria is not applicable and a fatigue analysis is required. Therefore, if the component were cycling every 5 minutes, and your design life were greater than 9.5 years, then you would not be permitted to use thee screening criteria.

Furthermore, the criteria for Method A is very very conservative. Your total number of cycles is limited to a maximum of 1000 (less if you have less-than-desirable features). You might pass a Method B (5.5.2.4) screening.

 

[XL83NL]

Yes I did understand that. The point Im trying to grasp here is that full range doesnt say anything about a structure's fatigue resistance, but more about the defitions a designer applies (whats operating, whats design), from what I can tell here.

I think the approach in EN 13445 on this fatigue assessment is different, as it looks (first) if any (pressure) cycle is greater than a certain threshold of the equipment's MAWP.
Let's consider this. I have a (hypothetical) vessel of x,y,z, dimensions and made of material M grade A, with:
- a design pressure of 50 bar(g), operating pressure of 35 bar(g), and MAWP of 65 bar(g), or I have
- a design pressure of 50 bar(g), operating pressure of 35 bar(g), and MAWP of 465 bar(g), or I have.
- a design pressure of 450 bar(g), operating pressure of 35 bar(g), and MAWP of 465 bar(g).

The latter 2 obvisouly are not realistic but possible. When looking at the MAWP vs design & operating, especially when those 2 are not so close, Im not sure if the 'full range' phrase really makes sense.

 

[TGS4]

XL83NL - these screening criteria are intended to be very conservative. So, getting into minutia of exactly what constitutes "full" vs "partial" is not necessary. In fact, for Method A, any cycle greater than 20% (or 15% for non-integral construction) of design pressure is counted the same. Method B handles them separately.

Now, once you get into a fatigue analysis, then you have to consider everything - and the insignificant variations in stress fall away due to Miner's Rule.

 

[XL83NL]

Hi Trevor, thanks for clearing that up.

 

[DSB123]

XL83NL,
You stated "Say I design some pressure vessel for 0-50 bar(g) design pressure." So your Design pressure is 50 Barg is it not!!! Then you say the pressure goes from 0 to 45 to 0 so that is a significant pressure cycle and must be counted for the screening. We know it is well below the MAWP of 248 Bar(G) but that just means you have provided excess thickness and not a good design!!!

 

[XL83NL]

The pressure vessels I deal with are usually small and have an MAWP that's typically a factor of the design pressure, because minimum (stock) thicknesses are used which provide far greater allowables. Or the design is so unique that other factors control the thickness.

I'm not in the usual market when it comes to pressure vessel design.

 

[BradyJr]

Also, does anyone have any recommendations for fatigue analysis software?

 

[TGS4]

I use my FEA software to calculate the stress ranges and then Excel or Mathcad (or even pencil/paper).

 

[BradyJr]

TGS4,

I'm new to fatigue analysis of pressure vessels so please excuse my ignorance. What section of the code are you performing with hand calcs for fatigue analysis? Are you referring to ASME Sect VIII Div 2 - Part 4, Part 5.5, or else where?

 

[marty007]

I assume TGS4 is mentioning hand calcs for post-processing of the FEA. Depending on your analysis style (elastic / elastic plastic...), you need to convert the FEA generated stresses into code-speak (eg. cumulative fatigue damage).

 

[TGS4]

You'll note, that I said FEA then Excel/Mathcad/hand-calc. My fatigue analyses are to 5.5. marty007's response is correct.

If you're new to the topic, I would recommend a training course in the topic.