Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Sizing Relief Vent 2

Status
Not open for further replies.

trojanRik

Chemical
Mar 26, 2015
3
0
0
US
Hello,

My group is designing multiple rupture disc relief vents and have been discussing with the owner/client the appropriate degree to which the vents can be reduced for cost savings. Using the resistance to flow method for determining the flow capacity of the vent + fittings + rupture disc, there is inherently a minimum line size that will allow sufficient flow at our inlet and outlet pressure conditions, but we are struggling to find definitive guidance for what velocity is acceptable in the vent.

Our typical velocity recommendations for gases in ducts would be 1500-2000 fpm. The vent sizes that the client is requesting would result in significantly higher velocities, in the 50,000fpm ballpark.

Can anyone please advise if there are regulations or references that would help give guidance? We have performed some thrust calculations to at least ensure that the high velocities would be tolerable by fittings and supports.

Any insight is appreciated.
 
Replies continue below

Recommended for you

What is being protected from overpressure? What is the design code of the vessel?


*** Per ISO-4126, the generic term
'Safety Valve' is used regardless of application or design ***

*** 'Pressure-relief Valve' is the equivalent ASME/API term ***
 
Have you checked sonic velocity?

And you will hear it in the next state.....

50000 fpm is about 250 m/sec. that's smoking....

but Regs? Probably not codified to that extent, but as said, think about the noise....

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
 
Relief vent line sizing is typically less about adhering to a given velocity limit (driven by noise, pressure drop economics, erosion, etc.) and more a question of the resultant pressure drop not impacting relief capacity.

Usually rupture disc vent lines to atmosphere that are reasonably short are just the same line size as the rupture disc/holder without issue, as the flow resistance of the rupture disc typically has a much larger impact on the relief capacity of the whole line than the downstream vent piping.

That being said, 50,000 ft/min (833 ft/s) sounds quite extreme and possibly choked/sonic.
 
thankyou all for the replies.

Some of our non-flammable systems are being protected from a failure in nitrogen blanket regulation. For the solvents/flammables, we are basing an overpressure scenario off a fire event in accordance with API.

Some of the proposed flows are "only" in the 10,000fpm ballpark, but the best references we can find say that 6,000fpm is the fastest velocity recommended for steel duct.

One of our other concerns is that for the corrosives, which use CPVC ducting, what pressures the socket joints can handle. Discussing with the CPVC vendors they state that the socket welds are as strong, if not stronger, than the ducting itself. From a pressure standpoint, I think the elbows and fittings can tolerate the associated pressures and thrusts of these high velocities. To many or your points, yes this surely will be extremely loud.

 
250m/sec is close to the sonic limit of about 300m/sec for most gases; we normally set up vent lines based on developed backpressure, and at the limits of permissible backpressure, we typically see peak line velocity no more than 40-50% of sonic for short vent lines.
 
Note that text below does not relate to dispersion of relief, only to transporting from a relief device to a destination point.

At a local refinery during a depressuring of a diesel hydrotreating unit an excessive vibration threw a DN600 flare header from a piperack and ~100m segment fell down to the ground but did not rupture.

Total's std. GS EP ECP 103 said:
13.5 Flare and cold vent lines sizing criteria
13.5.2 Lines downstream relieving devices, flare and cold vent headers and sub-headers
The design of these lines shall comply with the following:
• Minimum line size 2".
• Back pressure to be compatible with the installed relieving device
• Velocity and ρV[sup]2[/sup]

• Monophasic (gas)
- Intermittent flow (peak short duration/infrequent flow)
. Lines downstream relieving devices and sub-headers: 0.7 Mach maximum and ρV[sup]2[/sup] < 150 000 Pa considering the maximum capacity of the relieving devices even if this figure exceeds the actual maximum flow rate due to process limitation and the relevant occurrence
. Headers: 0.7 Mach maximum and ρV[sup]2[/sup] < 150,000 Pa considering the maximum flow rate due to process limitations and for the relevant occurrence, however a velocity of 0.8 Mach could be accepted for a long straight line without elbows and connections (e.g. stack, line on bridge).
For ρV[sup]2[/sup] values greater than 100 000 Pa, vibrations and line support studies are required.
For ρV[sup]2[/sup] values greater than 150 000 Pa but below 200 000 Pa, derogations can be granted on a case by case basis.
- Continuous flow (flow of longer duration or more frequent) Mach < 0.35 and ρV[sup]2[/sup] ≤ 50,000 Pa

• Multiphase (2 phase flow at the inlet of relieving device)
Mach < 0.25 and ρV[sup]2[/sup] ≤ 50,000 Pa for intermittent or continuous flow

BP's std. GP 44-80 said:
8.2. Design and construction
u. Relief piping shall be screened for AIV failure using the methods described in GN 44-005.
The use of sound power levels as described in GN 44-005 is the most reliable method of screening for AIV. Flow velocity, pressure drop change, or other parameters are not as reliable in identifying AIV issues.
v. Relief piping should be sized to limit the maximum velocity to less than or equal to Mach 0,8 to avoid sonic choked flow. Higher Mach may be allowed only if:
1. AIV screening analysis is performed.
2. Piping and pressure relief devices are adequately supported/braced for the reaction forces caused by venting.
Appropriate consideration of AIV or FIV is particularly important if connecting large diameter (greater than DN 200 [NPS 8] outside diameter), thin wall piping (Schedule 10 or Schedule 20) to flare/vent headers.

Chevron's std. 1900 said:
1913 Relief Valves Discharging to Atmosphere
Design Considerations
In addition to complying with local and environmental requirements, relief valves discharging directly to the atmosphere should meet the following good fire and safety practices:
...
• Per API 2510A, the velocity of discharge at rated capacity should not exceed 100 ft/sec.

ExxonMobil std. XVD said:
Acoustically Induced Vibration Problems in Header Systems
Large headers inherently have the potential for flow induced vibration problems. In particular, flare headers for gas piping systems in which high capacity pressure reducing valves discharge have experienced problems of fatigue failure where excessive turbulence and high acoustic energy existed. The turbulent forces excite complex modes of vibration in downstream piping components. These vibrations can in turn result in stresses exceeding the endurance limit for the materials and thus, fatigue failure. Pressure relief devices may have the capability of generating sufficient acoustic energy to cause fatigue failures in downstream discharge laterals and/or header piping.
Potential vibration problems of this type should be considered early in the design stage of the header system. Guidelines for assessing the potential of acoustically induced piping vibrations in pressure reducing systems are contained in the Piping Vibration Evaluation Guide, EE.21E.89 (Section 7).

1. The following screening criteria should be used to recognize services with potential vibration problems requiring further detail evaluation:
a. Downstream line size 16 in. (400 mm) and greater: mass flow rate greater than 200,000 lb/hr (91,000 kg/hr) or upstream to downstream pressure ratio greater than 3.
b. Downstream line size 8 to 14 in. (200 to 350 mm): downstream line velocity greater than 50% sonic and upstream to downstream pressure ratio greater than 3.
c. Downstream line size less than 8 in. (200 mm) swaged up or “Teed" to 8 in. (200 mm) or larger line: downstream line velocity greater than 50% sonic and upstream to downstream pressure ratio greater than 3.
The above criteria are a guide for detecting potential problems with gas letdown systems and apply for the piping downstream of the pressure reducer under concern. Systems with only liquid flow are not identified as potential problems and need not be investigated. For systems with two-phase flow, use the conservative assumption of the total mass flow rate as gas. Any system exceeding these criteria should be further evaluated in accordance with the guidelines and calculation procedures set forth in the Piping Vibration Evaluation Guide, EE.21E.89 (Section 7). It is further recommended that the Mechanical Engineering Services Section of EETD be consulted when a piping system problem is suspected.
A distinction must be made regarding the length of service of the pressure reducing systems. Fatigue failure of any mechanical system depends on time, i.e., the number of cycles to failure. Therefore, the treatment required for a continuous service may not be justified for a short-term service.

2. Short Term Service - A system in short-term service is defined as one which operates a total of 12 hours or less during the life of the plant. Pressure relief devices typically do not exceed this limit. Systems in short-term service exceeding the screening criteria indicated above should be evaluated in accordance with EE.21E.89. Services determined from this evaluation to require special treatment should be identified as follows in the Design Specification:
Notes for Design Specification - The following design features should be applied to the discharge piping and header for approximately 300 ft (90 m) downstream from the PR device in question. (A more rigorous distance formula based on acoustic power level is given in EE.21E.89.)
a. Use pipe with a minimum wall thickness of 1/2 in. (13 mm) to increase flexural stiffness.
b. Use completely welded full wrap-around reinforcement pads at branch connections per ASME B31.8, Figure 13, Sleeve Type, with pad thickness equal to header wall thickness.
c. Use wrap-around reinforcement at welded support shoes and anchors. Alternatively, all welding of these fittings to the pipe wall may be eliminated by the use of bolted shoes and anchors.
d. Minimize all vents, drains, and small diameter connections. Those remaining must be double gusseted.

3. Continuous Service - Pressure reducing valves which will be operated more than 12 hours during the life of the plant should be considered to be in continuous service. Such systems, which exceed the screening criteria given herein, should be further evaluated in accordance with EE.21E.89. Systems in continuous service believed to be fatigue prone per EE.21E.89 require more positive action to reduce the acoustically induced vibrations because of the greater potential for fatigue failures in these systems. Treatment alternatives for these services typically require measures to reduce the acoustic energy generated at the source. The Mechanical Engineering Section of EETD should be consulted when problems of this type are suspected.

4. Maximum Line Velocity - Sonic conditions at piping discontinuities such as at branch connections, reducers, etc., can also result in unacceptable acoustically induced vibrations. Maximum vapor or mixed phase flow velocities in piping should not exceed 50 percent of sonic for releases expected to exceed 12 hours during the life of the plant or 75 percent of sonic for releases having a lower cumulative duration (such as pressure relief valve releases).

Design for Startup Conditions - Closed headers must be designed for any abnormal conditions that may exist during commissioning of the header or plant startups.

Fluor Daniel's sec. 7.0 said:
7.2.2.4 Depressuring Lines
...
Where depressuring lines have been connected to the flare header, the results of flow in these headers must be accounted for in the back pressure calculations. If the potential flows through the depressuring devices are larger than the relief devices on the equipment protected, then consideration needs to be given to substituting the depressuring flows for the relief flows on a case by case basis. Where the size is 12” or under the velocity may be allowed to approach sonic. For larger sizes reduce velocity to 0.8 mach or less. These limitations have been established to minimize vibration and noise. Where the velocity is limited to 0.8 mach, perform hydraulic calculations with the outlet pressure at the low or no flow pressure in the relief mains (worst case for velocity).

Foster Wheeler sec. 603 said:
2.3 Header and Subheader Sizing
2.3.3 Suggested Velocities and Sonic Flow
...
Even if sufficient pressure is available, depending on the lowest set pressure of relief valve in the system, it is not desirable to size the header so that the flow becomes sonic (high noise level and pipe
vibrations).
To prevent this, the value for the ratio (Pz/P0)/(G/Gci) must be larger than 0.6 and preferably not less than 1, P2 being the pressure in the downstream vessel (See Lapple Chart).
The recommended range of values for flare headers and other piping when pressure drop is not controlling are listed below.
TABLE 3
RECOMMENDED HEADER VELOCITIES
<(P2/P0)/(G/Gci)> <V/V sonic>
Downstream piping of safety valves <1 to 2> <0.25 to 0.65>
Flare headers <1 to 2> <0.25 to 0.65>
Flare stacks <2 to 3> <n/a>

Norsok's P-001 said:
6.5 Sizing of flare and vent lines
6.5.1 General
In general, all flare lines shall be designed to keep the ρV2 < 200 000 kg/ms2 criteria (where ρ is the fluid density or mixed density for two phase conditions in kg/m3 and V is the velocity in m/s).
Further, the selection of piping specification shall consider the effect of acoustic fatigue, which is affected by factors such as
• relative differential pressure in upstream restriction device,
• temperature in the flowing gas,
• mole weight of flowing gas,
• pipe diameter and wall thickness,
• mass flow rate.
6.5.2 Flare headers and sub-headers
Piping for flare and sub-headers shall be designed for a maximum velocity of Mach 0,6.
6.5.3 Pressure safety valve lines
The upstream and downstream line shall be sized according to requirements in the relevant pressure relieving design code.
Maximum flowing velocity in the lines downstream of the PSVs to the first sub-header, shall in general be less than Mach 0,7. For the PSVs where the outlet velocity is higher, a reducer should be installed as close as possible to the PSV to increase line size and hence limit the velocity to maximum Mach 0,7 downstream of the reducer. Nevertheless, the actual back pressure at the PSV outlet and in the block valve shall be checked to be consistent with back pressure limitations.
6.5.4 Controlled flaring lines
Flaring lines downstream of control valves shall be designed for a maximum velocity of Mach 0,5.
6.5.5 Depressurisation lines
The maximum flowing velocity in the lines downstream the reducer shall be Mach 0,7.
The pressure loss shall not impose any restrictions on the depressurisation objectives.
6.5.6 Relief lines with slug/plug flow
For potential slug/plug flow, line sizing shall be based on slug velocity and slug density. These slug characteristics shall form the basis for stress calculations and design of piping support.
6.5.7 Vent lines
Maximum backpressure shall be 0,07 barg.

Samsung's std. SEM-9437E said:
2.3.2 Piping Arrangement
(1) Flare headers and subheaders: maximum velocity of 0.6 Mach
(2) PSV discharge lines, depressuring line: maximum velocity of 0.7 Mach
(3) Two phase relieving for potential slug/plug flow: V < 50 m/s
Where these criteria cannot be be met, additional calculations might be required to document that the selected pipe size is still acceptable. This involves evaluating piping stress levels, supporting, noise etc. All these criteria should not be applied for corrosive, erosive system.
If rV[sup]2[/sup] > 200000, the piping discipline shall be consulted in order to consider reaction forces(r is fluid density in kg/m3 and V is velocity in m/s for gas, liquid, two-phase).

Shell's std. DEP 80.45.10.10 said:
3.7.6 Flow Induced Vibration and Acoustic Induced Vibration
Relief system laterals and headers with flows from PRVs do not require a Flow Induced Vibration (FIV) study. Relief system laterals and headers with flows from other valves (e.g., a large pressure control/let down valve) require a FIV study where both 1) the flow exceeds the ρv2 criteria (see table below) and 2) this flow is intended to occur for extended periods of time (weeks).
image.png

An Acoustic Induced Vibration (AIV) study is not required for flare headers that are DN 600 (NPS 24) or smaller provided that standard wall pipe is used (schedule 20 or heavier).
For larger flare headers, if diameter/wall thickness > 70 then an AIV study shall be performed for any PRVs that produce more than 155 dB.

Technip's std. PEDG P4 S4 SS4.1 said:
5. FLARE LINES
Discharge lines of safety valves, flare headers and sub-headers shall be sized according to three simultaneous criteria:
- Pressure drop
- Fluid velocity
- Kinetic energy of the fluid
Pressure drop shall be limited by maximum allowable backpressure allowed at relief valve discharge.
Velocity shall be limited to 0.3 to 0.5 times the critical velocity of the fluid, which can be calculated using the following formula, based on ideal gas hypothesis:
image.png

Pipe fittings generate turbulence the impact of which is the reduction of the effective area and the increase velocity , which the risk of flowrate limitation if sound velocity is reached.
Flare system sizing is the subject of a specific design guide and of a dedicated computer program ...

Chiyoda's std. LTC-PB-P0-004 said:
11. SIZING OF FLARE HEADER
g. Limit the Mach no. of 0.2 (as per API 521) at the flare header.
image.png

IPS's std. IPS-E-PR-440 said:
13.5.4 Flare headers
Flare headers shall be designed so that the maximum allowable velocity does not exceed 50 percent of critical velocity, a figure mostly practiced by design companies.

JGC's std. 210-120-1-41 said:
4.2.2 Outlet Piping for Pressure Relief Devices
...
(7) Noise and vibration
Acoustically induced vibration shall be taken into the design of pressure relief devices in high capacity pressure reducing gas service. The evaluation of acoustically induced vibration shall be subject to consultation with the noise control engineer.
(8) Sonic velocity
For intermittent services, such as for pressure relief valve discharge piping, 80% of sonic velocity may be acceptable. ...

WorleyParsons' std. PTD-DGS-105 said:
The following definitions pertain to flare header sizing:
...
5. Mach number limit is the largest ratio of fluid velocity to the speed of sound in the fluid. Use a mach number limit of 0.5 for header and 0.5 for branches.

WorleyParsons' std. WS-DG-PR-007 said:
4.5 Discharge piping
...
The rated flow rate in the discharge piping should not exceed 0.75 Mach.
 
CCPS's Guidelines for Pressure Relief 2nd ed. said:
7.3.4 Vibration
Discharge piping from relief devices and collection headers is also subject to vibrations induced by high velocity discharges within the piping and from equipment connected to the piping. Vibrations can be induced by acoustical forces (high frequency), by internal turbulence (low frequency, usually less than 100 Hz), or by external mechanical forces (reciprocating compressors). Vibration can cause discharge piping and header failures, particularly in thin-wall piping or near small bore connections to larger piping. Failures due to acoustical forces (acoustic induced vibration or AIV) occur rapidly (in seconds) while those due to internal turbulence (flow induced vibration or FIV) occur after longer exposure times (minutes or hours). Failures due to external mechanically induced vibrations (MIV) usually require even longer exposure times (days). All such vibration-induced failure of relief collection header is harmful. The principal distinguishing feature is the duration between the onset of vibration and the header failure.
There are few recognized and generally accepted standards for prevention of piping failure caused by vibration. One frequently cited is “Guidelines for the Avoidance of Vibration Induced Fatigue Failures in Process Piping (2nd Edition, January 2008)” published by the Energy Institute in London.
Some companies require investigation of the potential for FIV in applications for which peak velocities exceed 50% of sonic. Some companies limit the potential for AIV failures by maintaining the sound pressure level below 157 dB for carbon steel piping. This value can be placed into perspective by observing that a sound pressure level of 168 dB is produced by 100 tons per hour of a gas with a molecular weight of 22 at a temperature of 50 C (122 F) experiencing a pressure drop of 80 bar (1175 psi). Other companies limit the product of gas density and the square of linear velocity to a threshold value (ρu[sub]g[/sub][sup]2[/sup] < 200,000 kg/m/sec[sup]2[/sup])(134, 146 lb/ft/sec[sup]2[/sup]).
Angled entries of laterals into flare headers should be considered during the design phase to minimize pressure losses and to reduce the potential for piping failures due to turbulence or acoustic vibration. The additional difficulties caused during construction almost always provide greater benefits in reductions for the potential for vibration induced failures and the potential for pressure losses during emergency relief flows.
 
Status
Not open for further replies.
Back
Top