PSV location 3

[PBFSI]

When designing vessel/piping arrangement, is there a rule of thumb as to where the PSV is located. i.e. on the vessel or as close to the vessel as possible? Anybody know the code requirements for this? Vessel is ASME Sec VIII div I, 2002 addenda. Any help would be very much appreciated.
Cheers.

 

[unclesyd]

The location used to be in UG-134. This is good if the code hasn't changed in the 2002 addenda

It can be on the vessel itself or on the vapor line if the vapor line and nozzle in the vessel is at least as large as the inlet of PSV and not so far as to cause a reduction in flow to the valve. The can be no stop valves in the line.
It is better to put them on the vessel proper as you get fewer questions from the jurisdictional authority. If it is a little distance from the vessel have you calculations handy. Having said all of the above there are circumstances that will allow stop valves in the line or before the valve. These are watched very closely by the jurisdictional authority as to their necessity. If you think stop valves are necessary get the ok before installing them. This information used to be in Appendix M.

We had some vessels, reactors, that the AI required the PSV to be on the vessels.

 

[KenA]

Check API-520 and API-521. PSV can be on vessel or outlet piping provided that (a) there is no restriction to flow which would reduce the PSV relief rate and (b) the total pressure drop between the vessel nozzle and the PSV inlet flange is <3% of the valve set pressure. However API-521 does also go on to say &quot;The pressure relief valve should also be located as close to the source of pressure as is practicable&quot;.
k

 

[NiravK]

In addition to what KenA suggested, for most cases in petroleum refining, PSV will always be located above the height of the Pipe Header(Flare Header) into which it is
discharging.
In other words, piping is required to have 'Free Draining' on both the side(inlet and outlet) of the PSV. If this is not considered, you may have liquid accumulation in the PSV connected lines which will creat back pressure and will not allow PSV to operate as per design conditions.

Regards,
Nirav

 

[PBFSI]

Thanks very much guys. The actual vessel is a bag type filter with a top swing-away lid. There is a 2&quot; NPT port near the top of the vessel shell. We obtained an off-the-shelf PSV set to the proper specs but it interferes with the vessel lid when opening. So I want to pipe-fit the PSV down to a better location... FYI
again, thanks
PB

 

[ixchawla]

I think all is said on the original question, which is available in codes API-520,521.
I have another query. What if the PSV is provided for derating the piping, when no vessel is there. How the following is taken care in design:
1. If PSV is located far-off from the derating point, is it possible to set the psv set pressure as pressure at derating point minus line loss at relieving conditions.
2. Does the 3% rule apply for such cases?

Thanks in advance.

 

[PAN]

From KenA's comment, The pressure relief valve should also be located as close to the source of pressure as is practicable".

From nirav2009's comment, PSV will always be located above the height of the Pipe Header(Flare Header) into which it is discharging.

In my case, the top elevation of vessel is lower than the flare header. What is the suitable location of PSV? Please comment.

 

[jsummerfield]

nirav2909 beat me to the punch when he stated "... for most cases in petroleum refining, PSV will always be located above the height of the Pipe Header(Flare Header) into which it is discharging."

Never say Always or Never. Steam, air and such atmospheric safety valves do not need to be above the flare header.

John

 

[roker]

hi,

if the psv is located below the flare header you have to connect a drain line (3/4") from the low point into a close drain system (as indicated in API 520/521).

Regards,
roker

 

[itdepends]

For interest- I read an article a while ago in an online magazine which suggested that in some cases it is better to install PSV's on the side/bottom of vessels.

The logic is based on over pressurisation scenarios where the vessel is a reactor and the pressure is being generated by a chemical reaction. For many reactions the rate of reaction increases with increases in temperature- which means that a reaction can run away from you (if it is exothermic).

In this case it is worth considering a PSV lower down the vessel (or on the bottom) to relieve pressure and remove the reactants (the source of the overpressure) at the same time.

Of course this is dependant on your individual application.

After all

ItDepends.

 

[brainstorming]

I think there is no stringent requirement in the location of the PSVs and from experience in the 3D model of some projects many PSVS have been relocated based on maintenance as well as operation views and needs.
Beside that, as a process engineer u have to check API-520 and API-521 requirements listed by KenA "(a) there is no restriction to flow which would reduce the PSV relief rate and (b) the total pressure drop between the vessel nozzle and the PSV inlet flange is <3% of the valve set pressure"

in some cases like Cooling water (non hazardous material) PSVs are located near grade to discharge water to the grade.

Regards

 

[121202]

PSVs do not have to be directly flanged to the vessel being protected. In fact this is rarely done given the general requirement that PSVs have to discharge to a flare system.

I'd give the following advice on layout and design of piping

- The PSV tail pipe should fall all the way towards the flare header and should be installed without pockets to ensure that the line is free draining

- The PSV inlet line should be designed so that it falls back to the vessel being protected without pockets so that it is free draining

- Permanent access (platforming & ladders)should be provided to PSVs to enable easy access to the isolation valves on the inlet and tail pipes

- The inlet line should be at least one pipe size larger than the inlet flange to the PSV. Reducers should be located directly upstream of the PSV.

- The outlet line should be at least one pipe size larger than the outlet flange of the PSV. Reducers should be located directly downstream of the PSV.

In terms of process piping design the following general requirements should be followed

- The pressure loss on the inlet line should be limited to within 3% of the set pressure

- The tail pipe should be sized to limit velocities to 0.7 x sonic velocity

- The piping\flare design should limit the back pressure (built up + superimposed) to the maximum limits recommended by the PSV supplier.

 

[brainstorming]

well done 121202
but with respect to backpressure, I experienced one event when I deleted some PSVs from a group of multiple PSVs since the flare header was revised and a new lower backpressure was calculated.
I could not understand that " The tail pipe should be sized to limit velocities to 0.7 x sonic velocity" is this a mandatory requirement. I feel it is not.


Regards

 

[121202]

Brainstorming,

I do not believe that the "0.7 x Sonic Velocity" requirement is a manditory code requirement. However, it is common practice to limit tail pipe velocities to this figure for two reasons:

- Firstly and most importantly, at velocities above 0.7 x sonic velocity there is a risk of acoustic fatigue in flare piping. This has led to the rapid failure of flare branch pipework in the past. Thin walled flare piping is particularly at risk.

- Secondly limiting velocity also reduces the noise produced.

Rgds

 

[Rich2001]

The 0.7 Mach number limit is to prevent transonic flow. Small variation in wall roughness, curvature, bends, abrupt changes inside valves and fittings may generate a local condition of Mach 1 flow (shock-wave) while still having an overall lower Mach number (<1). When transonic flow occurs, the drag increase is greater than would be estimated from a loss of energy through the shock. In fact, the shock wave interacts with the boundary layer so that a separation of the boundary layer occurs immediately behind the shock. This condition accounts for a large increase in drag known as shock-induced (boundary-layer) separation.

 

[brainstorming]

Rich2001
I respect your valuable explanation but this is more theoretically than anything else.
PSVs discharge flow are not continues service and might never occur in the plant life. So, considering your concern of mach number is not considered in the design otherwise it would (if I am right) cost much.

121202
Regarding the noise, I have experienced PSV discharge piping with an insulation (Acoustic) to reduce noise that because it is cheaper than sizing up the piping (cost saving) and the justification that these are intermittent services.

Regards

 

[Rich2001]

The point of the above post is at a mach number greater than 0.7 there is a high probalility that the flow can realize a localized mach 1 condition, id est choked flow.

A contributing factor to the fatal incident at a chemical facility, belpre OH May 1994, was that the PSV system was inadequate due localized choked flow.

 

[brainstorming]

Good to learn such incidents
Would you please explain what happened in brief

Regards