PSV SET POINT and ASME standards 3

[ashwani25]

Exceeding the design pressure as set pressure for the PSV for a ANSI 900(2220 psig) pressure system. What kind of problems can you face while in operation. following are the operating parameters.

Design factor for all fittings and piping is at 0.6 and materials adhere to API and ASME standards for carbon steel.
Product in service is Liquid C02 being pumped at max discharge of 2000 psig.
ANSI 900 pressure rating for stainless steel (PSV material) 2160 psig.
Current PSV set point is at 2260 psig.

My understanding is that this is incorrect as PSV set point should not be above the design pressure for the piping . Please provide your comments on what should be the correct method way to determine you the PSV relief set point for this case.

 

[BigInch]

Read your design code.
Codes often permit setting the PSV at up to 10% above allowable pressure.
Allowable pressure is the lowest design pressure of any component in the system.

you must get smarter than the software you're using.

 

[LittleInch]

ashwani,

I think your understanding is correct. BI is correct, but usually these allowances are only for occasional events and the design pressure should not normally be exceeded in a steady state operation. If you have other alarms or trips set below the relief valve set pressure but above the max operating presusre, this might be allowed on the basis that your relief valve is a last line of defence, not the primary trip / shutdown / protection system for the pipework. On that basis the occasional pressure blip rule might start to apply as you don't want the relief valve lifting before you reach the trip settings on your instrumentation. Alld epends on your system and it's various protection devices and how reliable they are.

The operational issue is that your max discharge and your design pressure are quite close and for a simple spring valve, the reliability of the relief valve is often in the range 8 to 10% below the set point when the valves can start to "chatter" and lift a little bit.

If you need tighter accuracy you probably need to go to a pilot operated valve which can often get down to the 3 to 4% accuracy range.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[ashwani25]

but exceeding the piping pressure class , isn't it a violation. I know you can set the PSV's at the pressure class for eg:- for ANSI 900 you can have the PSV set at 2220 psig but setting it at 2350 or such is violation of the piping codes . I know ASME B 31.8 tells to not go above 10% of the MAWP but these values are always less than the piping pressure class rating.

 

[LittleInch]

ashwani,

Exceeding the design pressure on a regular operating basis is a violation and if this relief valve was your only means of pressure safety then you are correct.

However, in many systems there are other safety devices (instrumented pressure trips etc) and may be the pressure cannot exceed the design pressure.

See this thread,

http://www.eng-tips.com/viewthread.cfm?qid=229469

and many others which discuss the interrelation between pressure limits. BTW I don't recall B 31.8 defining MAWP and it states clearly that design pressure is the maximum pressure allowed by the code. If you can point ot a specific clause or section then do so, but don't take anything said as fixed if they can't point to a specific clause. There is a great deal of confusion and misunderstanding unfortunately about different abbreviations and what they mean.

Other codes have different meanings and allowances for overpressure as stated above. Pressure vessels talk in terms of MAWP which is not normally applied to piping systems and they have different "rules" about what pressures you can and can't exceed for different periods of time.

Setting relief valve pressures is always an interesting discussion when you get tight on margins. Relief valves are great and have a very high reliability, but are not as accurate as instrumented systems and you often need a more risked based system to decide the likelihood of a particular set of circumstances happening to allow you to set the relief valve higher than the design pressure if your design code allows it - read the code.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[ashwani25]

ASME B 31.8, 2012 , Article - 843.3.4, Clause (a)

 

[LittleInch]

I've seen that before and think it's rubbish. This refers to STATION pipework (has DF of 0.6 ) and is in a subsection called "...in Compressor Stations". Therefore not applicable IMHO, to the pipeline. See section 845 for the pipeline, which defines MAOP.

You rarely see 31.8 design inside stations, but you've not stated what code you're working to.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[BigInch]

If the code is B31.8, you might want to read INTERPRETATION 16-16 of 13 Jan 2012
Question (2): Can gas pipelines be designed to have transient overpressures up to 10% above
MAOP, similar to ASME B31.4?
Reply (2): No.

Now my question is why are we looking at B31.8?

B31.4 covers Carbon Dioxide Pipeline Systems, assuming you are transporting at above it's critical point, where it is highly efficient to do so BTW. See B31.4 Chapter X

LittleInch: Actually it is not so rare (from a world-wide perspective) to see stations designed to B31.8 as in the US the design of gas pipelines and compressor stations must meet the requirements of federal law (CFR Title 49, Part 192), which is (essentially) .... a copy of [highlight #EF2929]B31.8[/highlight]

you must get smarter than the software you're using.

 

[LittleInch]

BI - Thanks for that interpretation reference - Will be useful in the future. 3.18 was mentioned by the OP in one of his posts and I just latched onto it.

Fair enough about the US compressor stations, but do they not also use 31.3 for the piping or actually design piping to 31.8?

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[BigInch]

They don't use B31.3 ... or 31.8 for that matter.
CFR Title 49, Part 192 is used. (OK that's mostly B31.8)

In fact I've written to PHMSA to see if they will comment informally on the current practice in the USA. Until I hear back from them, if I do, please accept the following as prima facie evidence.

The attached is a Compressor station inspection checklist, that I downloaded from Pennsylvania's PUC, but you can download the same from the Fed's PHMSA website and other state's PUCs as well. As it is a Federal regulation, all states must abide equally. Note page 2, item .101
Question: [highlight #FCE94F]Is a design factor of 0.50 used if the compressor station is located in Class 1 or 2 areas?[/highlight]
That's not a B31.3 derived requirement, as you can tell by the specific use of DF and Classes.

If you used B31.3, you'd have to tick the [NO] box. B31.3's equivalent DF in Compressor stations is often, if not always, slightly higher. At least for the pipe walls and grades that I've checked recently.

I also hope you like the checklist as well. No doubt somebody you know will find it useful. I suppose it could even be adapted for B31.4

you must get smarter than the software you're using.

 

[BigInch]

I decided to check the older interpretations and good news, there is an old, old clarification (still not older than me)dealing directly with this issue. Interpretation 6-6 of 1989. Seems to leave no doubt (in my mind for sure) that B31.8 applies to piping within CS. Type C construction (now defined as Class 3 Area piping) is applicable from the station suction flange to the station discharge flange, exclusive of the compressor unit piping (which I have always known to be interpreted as that piping supplied by the vendor as part of the compressor itself, flanges, balance lines, etc.).






you must get smarter than the software you're using.

 

[BigInch]

An interesting variation was also discussed in one of the other Interpretations I saw. The question was, "What design factor is required for a pipeline passing through a compressor station yard, but not actually connected to the compressor station itself?" Presumably this came about when a line in a Class 1 or 2 Area is paralleled by another pipeline which actually required a compressor station and the CS land encompassed both pipeline's easements. The pipeline connecting inside the compressor station requires a DF of 0.50 (from station suction to station discharge flanges), but 0.72 outside the flanges and for the other parallel pipeline just passing through the station property, 0.72 was also sufficient.

you must get smarter than the software you're using.

 

[BigInch]

Here it is,

you must get smarter than the software you're using.

 

[sanshu1111]

As per the requirement of ASME Sec VIII Div 1, PSV set point (start opening) can not be more than MAWP in case of single relief device. It shall prevent pressure from rising more than 10% or 3 psi (which ever is greater) above the MAWP.

 

[BigInch]

Exactly why I said in the first post to read the particular design code. All we really know so far is it is probably B31.8
BPV VIII is for vessels (possibly within a pipeline), but not specifically for pipelines, therefore we can assume (for now) that it is not applicable here.

you must get smarter than the software you're using.

 

[BigInch]

I have received a response to the question I raised with DOT/PHMSA concerning the use of B31.3 inside pump and compressor stations. In short they stated that

1) The requirements within the USA for DOT regulated pipelines (and compressor & pump stations) are the CFRs Title 49 PARTs 192 & 195
2) Agreed that these are similar to B31.4 and B31.8,
3) Have confirmed that B31.3 does NOT apply to pump and compressor stations.

They did however also state that B31.3 could be used within a pump or compressor station, just as any company standard, or even any totally arbitrary provision could, as long as it was applied as a supplementary requirement, with the explicit provision that the resulting design still meet all the requirements of CFR Title 49 Part 192 / 195 (B31.4 or 8).

 

[Duwe6]

Thanx, BigInch!