Safety/Relief valve Testing program 3

[MJCronin]

To all,

What is the best source of information to get specific infromation on a plant-wide safety and relief device testing program ?

I am aware that OSHA has some general "vague" requirements (OSHA 1910) that a plant owner must be able to show records of testing. But what document puts these devices in different categories of importance ?

What guideline requires more frequent testing for some devices as opposed to others ? Who has a website or refernce that I can turn to ?

Thank You in advance !!!

MJC

 

[BillBirch]

Good question,
This issue rears it's head every couple of years. In my experience in offshore south-east Asia, the governments used to allow the majors such as Shell and Esso to self-regulate their inspection and testing programmes. With regular shutdowns and loss of production, management were regularly revisiting the ctiteria for vessel internal inspection frequencies and RV recalibrating frequency. The general reaction to these rquests was to look at what eothers do, find that you do that same, and change nothing!
Sorry if this isn't very helpful, but seriously I don't know of any guidelines other than high level "safe-guarding consideration" statements, but I guess the old criticality, service, temperature considerations would be good "first principals" starting points. I don't think you will get much joy from valve manufacturers on valve reliabiliy.
Regards,
Bill Birch

 

[kstaylor]

The documents API-510, Pressure Vessel Inspection Code, and API-576, Inspection of Pressure Relieving Devices, have some good guidelines for setting of inspection intervals on these devices as well as how to inspect and what to inspect for. These are both available from The American Petroleum Institute (

http://www.api.org).

 

[marah]

I am working in an oil industry ,I need to know the following things
Popping Pressure of the safety Releive Valve
Set point
seat tightning pressure

what are the main tests for Pressure Safety Releive Valve?

 

[TD2K]

Marah, look at this link:

http://www.andersongreenwood.com/docnet/pressprd.asp

Download the Crosby engineering book and go through it, it's a good reference.

 

[svanels]

We are looking for practical information to make our own psv test bench. Any information on this subject is highly apreciated.

Regards Steven van Els
SAvanEls@cq-link.sr

 

This is a very interesting subject, not only from the perspective of maintaining and testing relief devices, but for looking at your whole program. What about the pressure vessels these devices could protect ? Do you inspect these? Is your relief device properly sized? How do you know? Have you investigated all the senarios? Another good source for starting a program would be the NBIC (National Board Inspection Code). Is your state governed by ASME? Are you a code state?
From my experience a comprehensive program could take a long time to establish. We size each relief device for the worst case, check set pressure before installation on a test stand, pull the device after one year, pop it and look at the as found condition. At this point we determine based on valves in like service and in the "as found" condition if the inspection interval should be changed. We keep a valve history on each position and have a maximum length the valve can stay in before it is pulled. Occassionally we will a valve that has an exception (pops high, dirty, etc.). This position now requires an investigation to determine why? Another area we look at with relief devices that is often overlooked is to check sizing when the process changes. A process change could change the relieving requirement! We also do external and internal visuals on our pressure vessels, and do some pressure testing as well.

 

[svanels]

We have a Test & Inspection Turn Around once in the 3 years. In this period we have the time to do our plant wide inspection on equipment that is always in service. Vacuum colomns, vessels, heaters, boilers etc., common instalations found in Oil refinery. A detailed schedule is worked out, an as part of this we cannot overlook psv's, trv's etc, or everything which could delay a startup.
Our plant is not in the USA, and for testing the maior valves we send them abroad. Steven van Els
SAvanEls@cq-link.sr

 

[kahlilj]

i work for a natural gas pipeline company. we are in the midst of re-evaluating our relief valve capacity calculations & possible resizing. we came across a couple of terms that seemed confusing if not inconsistent. in determining the capacity of a relief valve, API 520 has defined the pressure term in their equation to be equal to the set pressure (of the relief valve) plus the allowable overpressure plus atmospheric pressure. i understand that the allowable overpressure is 10%, but 10% of what? the set pressure of the valve (as i've seen in some AGCO sample docs) or 10% of the MAOP of the piping system (i.e. the pipeline)?? anyone know which should be used?

 

[TD2K]

It depends (don't you just hate that) though in reality, the accumulation refers to 110% of the MAWP (note, API-520 is directed towards pressure vessels, you just need to ensure the overpressure limits you use for your pipeline calculations meet the applicable pipeline code also).

I've seen lots of calculations (mostly for pressure vessels mark you) where the PSV may be set lower than the MAWP for whatever reason yet the calculations are still done at a relieving pressure of 110% of set pressure (116% for multiple PSVs, 121% for fire). Another example is we do our piping relief calcs at 110% accumulation although the B31.3 code allows 33% accumulation for short term excursions. If the EXISTING PSV doesn't have enough capacity, then I've seen a higher accumulation used as long as it is within the equipment code limits. For a NEW PSV, I doubt I'd be allowed to take credit for this, I'd be told to buy the next larger PSV. Anyway, back to the first part of this paragraph, if the vessel is designed for 500 psig but the PSV is set at 400 psig, its capacity calculations would normally be done at 440 psig (converted to absolute of course). However, as long as something else in the system wasn't rated for a lower pressure, you 'could' do the relief calculations at up to 37.5% accumulation. With a 400 psig set pressure, that would correspond up to 550 psig relieving pressure, 110% of the vessel's MAWP value of 500 psig (with similar logic if you had multiple PSVs, fire relief, etc). What you're protecting is the vessel and the allowable overpressure limits (or pipeline) and ultimately its MAWP is what you need to prevent from being exceeded in excess of the allowable amount. Our course, company's practices come into play and yours may not allow this arguement.

 

[MortenA]

kahlilj:

The 10% overpressure is calculated with respect to PSV set pressure and is for spring operated valves only. The 110% is for single service valves. Please note that if your relief case is "fire" then max allow overpressure is 21%.

The meaning of 10% allowable overpressure is that the max capacity is acchived at 10% higher pressure than SP, but the valve will actually start to open at SP (and "shimmer" somewhat before that).

A pilot operated valve achives full capacity as soon as the SP is reached.

Read the code carefully before attempting to size a valve. The figure where all the various pressures and % is marked on a scale first in the code should be very helpfull. Remeber all the small "pitfalls" such as pressure loss in header piping and back pressure in discharge piping!

But as TD2K mentions: This code is for pressure vessels and plant piping.

Its quite normal to select SP=vessel design P, since plant vessels and piping usually is tested to 150%.

The requred capacity is also calculated with respect to the 110% = smaller valve

Best Regards

Morten

 

[kahlilj]

thanks morten & td2k,
these relief valves are primarily for piping (although there will be some on vessels). if i understand your posts, i should be using an equation from ASME for relief valve capacity calcs on pipeline and API 520 for relief valve capacity calcs on pressure vessels??

anyway, i will look at these comments more closely and get back to you if i have more questions. your comments are greatly appreciated!

 

Is there an industry standard or best working practice that determines the frequency with which a PRV should be tested?

Does this requirement or practice determine the test method or is this left to best practice, given the application?

Any comments would be appreciated.

 

[kahlilj]

canadienPR,
our company tests RV's annually. the US's DOT states that these devices are to be tested every year not to exceed 15 months. that's what we comply with.
kgj