Low Pressure Vessel Pressure Protection

[Berenger]

Hi everyone,

We are installing a new low pressure 2-phase separator in the field. The operating P is around 120 psig. The pipeline and the vessel are to 1375 psig @ 185 F.

I had a discussion with another engineer on how best to protect this vessel. His point is that since the operating P is very low compared to the design, protection using PSV provide enough protection? His logic (and reality) is that we should not see a pressure above 300 psig (at worst) and the vessel is designed to 1375 psig (PSV also set at 1375 psig).

However, I feel that, even though we may never see a pressure over 300 psig, we also need high alarms just in case. We don't want the remote scenario of extremely high pressure to happen and we only have the PSV to protect us. If we have alarms, we will be able to respond to any unplanned increase in pressure before the PSV goes off. Also, what if the PSV fails? That leaves us with no layer of protection. Hence, the alarms will provide another layer of protection.

Am I correct in wanting multiple layers even though we realistically don't expect to use them? Or should we just go with only the PSV and cross our fingers that the never need it?

 

[GHartmann]

You need to do a mini HAZOP on your system.

What is happening upstream to cause the high pressure in your vessel? What is happening downstream to cause the high pressure?

Can any other equipment be damaged if your pressure exceeds your normal pressure of 120 psig and reaches 300 psig.

I have seen HAZOPs go both ways. Some insisting on one more layer of protection. Others did not consider the case viable and no extra layer of protection is required.

I am with the second group. We will see what others say.

 

[LittleInch]

I agree with GHartmann.

first you need to look at your pressure sources and work out what set of events could lead to the highest possible pressure. If that is still less than the design pressure then you have no issue to deal with. You need to decide if your "remote scenario of extremely high pressure " is actually feasible or not or whether it requires multiple failures of other systems pressure controls. Is there such redundancy / trips / valve closure on the lines feeding the separator with high pressure that you can take account of / credit for?

Your point "Also, what if the PSV fails?" fails to grasp the sheer reliability of simple spring loaded relief valves. These things virtually never fail. Their accuracy is sometime a bit low (-8%), but in terms of not going off their only problem comes if you've undersized it.

Do you have pressure transmitters on this vessel? If so just set a software alarm if you're that concerned rather than a specific high pressure switch / trip.

with high pressure pressure vessels you should never have to "cross your fingers" but instead deal with the issue in a rational and agreed way, then write it down logically and distribute it for comment / approval.

My motto: Learn something new every day

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

 

[moltenmetal]

If you have an ASME vessel, you REQUIRE a relief device, whether there are relief scenarios or not. It can be set as high as the MAWP of the vessel, or as low as you feel is necessary to protect the balance of the system from harms other than failure of the vessel itself, but you are not given the option of NOT having one unless the "vessel" is operated open to the atmosphere. As to what other layers of protection are required for your particular situation, you need to do that analysis with your group, investigating all scenarios in a systematic fashion (a "what if" or HAZOP review) as noted by others. That analysis can be simple or complex depending on what is in the process and the consequences of it getting out. And it should never be forgotten that a relief event is seldom a no-hazard/no consequence event. Relief events should not be routine- they should only occur when there is an emergency. Otherwise, you're challenging the availability of the device in a real emergency.

 

[bimr]

The function of a pressure relief valve is to protect pressure vessels, piping systems, and other equipment from pressures exceeding their design pressure by more that a fixed predetermined amount. The permissible amount of overpressure is covered by various codes and is a function of the type of equipment and the conditions causing the overpressure.

It is not the purpose of a pressure relief valve to control or regulate the pressure in the vessel or system that the valve protects, and it does not take the place of a control or regulating valve. The aim of safety systems in processing plants is to prevent damage to equipment, avoid injury to personnel and to eliminate any risks of compromising the welfare of the community at large and the environment. Proper sizing, selection, manufacture, assembly, test, installation, and maintenance of a pressure relief valve are critical to obtaining maximum protection.

ASME specifically states in Section VIII, Division 1, paragraph UG-125 (a) “All pressure vessels within the scope of this division, irrespective of size or pressure, shall be provided with pressure relief devices in accordance with the requirements of UG-125 through UG-137.” The pressure relief device may be the minimum size of 1-Inch for blocked-in thermal relief situation.

I agree with the others that you need to do a HAZOP on your system. The high design pressure does not make any sense in lieu of your proposed operating pressure. If you are correct, your company is throwing money away with the equipment over design.

Pressure relief valves should be designed to passively protect against a predetermined set of worst case conditions and should be installed to react to these conditions regardless of daily operation activities.

A credible worst-case scenario should be defined. For a given vessel, several plausible scenarios may exist – from external fire to various operating contingencies, such as overfill or vessel swell conditions. System overpressure is assumed to be caused by the controlling scenario. Most controlling scenarios are loaded with conservative assumptions that are never achieved in actual operating conditions. It is the controlling scenario relieving rate that dictates the pressure relief valve size. If sized correctly, the pressure relief valve should have enough discharge capacity to prevent the pressure in the pressure vessel rising 10% above its maximum allowable working pressure.

You need high alarms just in case of what? Alarm systems assist process operators in managing abnormal situations. One important alarm philosophy is that the operator must have some action for any specific alarm. If the action is not required, the alarm should not be installed. Improperly configured alarm systems make contributions to accidents. Nuisance alarms, alarm floods and improperly prioritized alarms all contribute to operator confusion, and thus increase accident frequency.

 

[MortenA]

@Berenger: Take a look at API 14C/ISO 10418 - this should give you a certain knowledge wbout what is needed. In addition to this look at the advice given above. You have to consider your whole system before deciding what is best.

As a first guess i would say: PSV with a SP= 1375 psig. If the upstream poressure CANNOT exceed 1375 psig for process reasons e.g. well shut-in pressure<1375 psig or MAWP upstream vessel <1375) then your design case could e.g. be fire. Once you have your system design according to recognised standards/RPs then do a HAZOP of the system and see where this leads you.

Best regards, Morten

 

[GarethChem]

I agree 100% with everyone else. You definitely require a PSV and should do a HAZOP. Even though you will never operate at 1375psi, you need to remember if there is a fire, the metal weakens and pressure increases.

PSVs are very reliable (0.01 - 0.67 faults/year for a 1/1 system,

http://tinyurl.com/o5rbl8q)

, however, if you want a second layer of protection, why not add a second PSV? One set to 600psi, and one set to 900psi? (these values obviously are dependent on the pressure ratings of the rest of your system)

You should have a PT on the vessel, so it should be pretty easy to program an alarm for a higher pressure at no extra expense (i.t.o. hardware).

Gareth

 

[MortenA]

It dosnt make much sence to install a 2. PSV. Actually i wouldnt consider a PSV as the primary protection - that would be your PSHH and associated ESD. The PSV would be the secondary protection. The PSHH should have a SP lower than the PSV (10%) in order to avoid PSV lift. The API/ISO std says it all.

Best regards, Morten