• Beta
Pressure Relief Devices (RV/RD/VB)
  • 09 Aug 2024
  • 7 Minutes to read
  • Contributors
  • Dark
    Light
  • PDF

Pressure Relief Devices (RV/RD/VB)

  • Dark
    Light
  • PDF

Article summary

Take Note
There are some specific cases where the S-RBI assessment approach in IMS PEI changes, depending on the Equipment/Component type.
This section gives guidance for Pressure Relief Devices (RV/RD/VB).

The functionality of relief valves can be categorized as:

  • Primary:
    • To avoid over-pressurization of associated Equipment and Piping.
  • Secondary:
    • To provide pressure relief capacity.
    • To separate the process from flare system (seat tightness).
    • To maintain pressure envelope integrity.
    • To lift to full travel when required and seal off when there is no demand (mechanical functionality).

Note: While the term PRV is sometimes used interchangeably with Pressure Safety Valve (PSV), there is a difference between the two. A PRV opens gradually in relation to the pressure, while a PSV is opened suddenly once the pressure hits a certain level in order to avoid over pressurization and a potential process safety incident. PSVs can be managed in IMS SIS (or SIFPro). 

Define Asset – Step 1

See Step 1.

Add a Component of Type "BODY" to the Pressure Relief Valve (PRV) Equipment. A Corrosion Loop always needs to be assigned to this Component. This can be the CL of the Equipment that the Relief Valve protects or all Relief Valves for a unit can be placed in one dedicated loop.

Graphical user interface, application  Description automatically generatedCreate Component Type BODY.

 

Start RBI Analysis and Assign DMs to Component – Step 3

See Step 3.

For this Component type, IMS automatically assigns Demand Scenarios, rather than Degradation Mechanism (DM), to the Component when it is created. You can also add the Demand Scenarios manually.

Rather than a DM, Demand Scenarios is assigned to the Component.

When you start the RBI Analysis, you first need to define the Demand Scenarios:

  • Click Demand scenario.
Graphical user interface, table  Description automatically generatedClick on Demand Scenario to define the Demand Scenarios.

 

  • Click + and fil in the Demand Scenario name and notes.
  • Click + to add more Demand Scenarios.

Determine Most Credible Failure Scenario – Step 4

See Step 4.

Define the Failure on Demand Scenarios.

For each Demand Scenario:

  • Select the Criticality Assessment tab;
  • Click on Demand Scenario;
  • Highlight the Demand Scenario; and
  • Click Scenario;
  • Click Edit.
  • Enter the Demand Scenario.Graphical user interface, text, application  Description automatically generated
    Write the Failure on Demand Scenarios.


StF – Step 7

See Step 7.

Methodology

Demand Rate 

  • Demand Rate reflects the number of times the PRV is required to lift and prevent over-pressurization.
  • Where possible real lifting information should be used.
  • The starting point to determine Demand Rates are the Relief Cases.
    • The safeguarding narrative or memorandum usually specifies as a minimum the 100% capacity relief cases.
    • A relief case generally starts with an “event” where the “event frequency” determines the Demand Rate.

Relief Cases are the scenarios that may cause over-pressurization of a system and affect the relief and flare sizing. Relief Cases are generally divided into three types (see the table below).

Relief Cases:

Relief Case
Explanation

Common Failures

Common failures are failures that can happen and are common to the unit or even the site. Common failures that may happen site wide, will – in most cases – determine the total flare capacity. Typical common failures are total instrument air failure and total power failure.

Fires

For the areas where a sustainable fire may develop, fire cannot be excluded as a case to lead to overpressure.

Single Failures

Single item failures are the most known causes for a relief. Single item failures can be caused by many actions or events.

Note: A well written safeguarding narrative/memorandum specifies all the Relief Cases and generally, the event frequencies for those cases determine the Demand Rate. For example, “heat exchanger tube rupture” scores in the “>20yr” category.

Likelihood of Failure

When determining the Likelihood of Failure, one should focus on the likelihood of the three Failure Modes causing a Relief Valve to “Fail to Danger” on demand (see the table below).

Failure Modes:

Failure Mode
Explanation

Sticking

Can it stick? Sticking by product, dirt or galling from both inlets, or outlet of the PRV:

Corrosive, solidifying material or polymers might cause the valve to stick to its seat, causing late opening or not at all. A well-known example in industry is the formation of magnetite in steam service, ‘gluing’ disk to seat.

Plugging

Can it plug? Can fouling, corrosion product or polymer prevent the PRV from properly functioning? 

A relief valve’s function is considered impaired when the diameter is reduced more than 20%. Condition of insulation and tracing should be considered here too.

Jamming

Can it jam? Can mechanical stresses, such as piping, cause valve jamming? Can, unqualified, maintenance, work cause mechanical failure in the valve?

Can the PRV fail due to mechanical faults or stresses? Are the inlet and outlet piping aligned or is the piping brought in place with brute force? Both hot and cold conditions should be considered when assessing stresses. Is maintenance of the PRV done by a qualified workshop technician? Are procedures in place and followed? Are original spare parts used?

Note: For pilot operated PRV’s it is suggested to use a minimum likelihood of “Low” because this type of PRV is more prone to plugging and/or fouling, due to the presence of small-bore tubing.

Confidence

The Confidence Rating reflects the confidence that the RBI team has in the Likelihood of Failure and depends on the stability of the process and the results of inspection/testing. The Confidence Rating is done for a specific PRV.

Confidence can lower the overall StF by one or two levels. Do not take credit for vents to safe location, as fouling, birds, and water accumulation cause issues. Cool service is valid if no temperature correction for test pressure is required (generally 100°C/212°F).

PRV Confidence Assessment Guidance:

Key Questions
Guidance

In service performance & as-received pop test performance

Both inspector and maintenance engineer should answer this question, as the question covers both regular inspection intervals and notification history.

If pop test results have always been acceptable and inspection did not reveal any fouling, debris or other blockage in inlet and outlet piping this question could be answered with Yes.

If there are no inspection/maintenance records or any abnormalities have been observed this question is answered with No.

Reliable in-service performance

Answer with No if there is a history of mechanical damage & repair (e.g., from chatter, bellow failure) or if operating within 90% of Set Pressure (SP).

Answer with Intermediate when passing in service. Note: Locations where flaring or emission regulations are in place, should score passing in service as No.

Yes scores the same as Intermediate for this question.

Cold and clean service

Known cool and clean service under all process modes, including flare or downstream side.

This question should be answered by both operator and technologist, as they have awareness of process conditions and possible fouling, polymerization and hydrate forming.

If the PRV is positioned downstream a bursting disc, this question can usually be answered with Yes, provided the outlet is proven to be clean and the inlet piping cannot plug.

PRV’s with relief to atmosphere score No by default, as outlet (piping) is prone to atmospheric fouling, plugged drain holes and bird nests.

If in doubt or not sure, chose the more conservative approach and answer the question with No.

Cool is any PRD for which no correction for temperature is required to calculate the test pressure (above 100°C/212°F, a correction factor is applied).

Clean service does not cause scaling, deposits, caking or plugging at all.

Overall StF

IMS uses the Demand Rate and Likelihood of Failure to lookup the overall StF (see table below) for each Demand Scenario. The Confidence rating is then used to correct the StF based on “proven reliable service” of the PRV.

StF lookup table for PRVs:

StF

Demand Rate

(>20yr)

(4-20 yr)

(0.5-4 yr)

(0-0.5 yr)

 

Likelihood of Failure

H

M

H

H

H

 

M

L

M

H

H

 

L

N

L

M

H

 

N

N

N

L

M

 

Software

Do an StF for each Demand Scenario:

  • Select the Criticality Assessment tab;
  • Click on Demand Scenario;
  • Highlight the Demand Scenario; and
  • Click PRV StF to start the StF assessment.
Graphical user interface, text, application, email  Description automatically generatedSelect a Demand Scenario and click PRV StF to start the assessment.

 

  • Select the Demand Rate (i.e., how often there is a demand).
  • Click Edit.
  • Select the Likelihood of Failure (to open on actual demand).
  • Answer the 3 Question (making use of the information buttons) to determine the Confidence.
  • Click Close.

Repeat to create more Demand Scenarios.

Consequence – Step 8

See Step 8.

Methodology

Remember to first write down the failure scenario (see step 4 above). Not all failures have consequences.

For External Fires:

  • Only assess incremental damage when the Relief Case is “external fire”; and
  • Take Boiling Liquid Expanding Vapour Explosion (BLEVE) and steam explosions into account.

The table below summarizes the significance and potential Consequence due to pressure accumulation. Accumulation is measured as % Over Maximum Allowable Working Pressure (MAWP).

Potential Consequence due to Pressure Accumulation (% Over MAWP):

Accumulation (% over MAWP
Significance
Potential Consequence

10%

Code allowable accumulation

No expected consequence

16%

Code allowable accumulation (multiple relief devices)

No expected consequence

21%

Code allowable accumulation for external fire relief

No expected consequence

50%

Common hydrostatic test pressure (may be 30% on new designs)

Possible leaks in associated instrumentation, etc.

90%

Minimum yield strength

Significant leaks probable. Failure of corroded/damaged vessel areas.

200 - 300%

Note: This value depends on code an material. The table used for this is standard pipe material.

Ultimate tensile strength

Catastrophic vessel rupture predicted.

Software

Do a Consequence of Failure Assessment for each Demand Scenario:

  • Select the Criticality Assessment tab;
  • Click on Demand Scenario;
  • Highlight the Demand Scenario; and
  • Click Asset, People, Environment, and Community, to assess each of these consequences.
Graphical user interface, application  Description automatically generatedClick Asset, People, Environment, and Community, to assess each of these consequences.

Note: Consequence of Failure on Demand is NOT identical to the RBI Consequence for the Equipment that the PRV protects.

Confidence – Step 10

See Step 10.

The Confidence is evaluated in the StF section (see Step 7 above).

IS and MII – Step 11 & Step 12

See Step 11 and Step 12.

The highest Criticality of all Demand Scenarios becomes the overall Criticality. IMS uses a lookup table to determine the MII from this overall Criticality. For Criticality class E a redesign is advised.

MII lookup table for PRVs:

MII

Consequence

N

L

M

H

E

Susceptibility to Failure (StF)

H

8

4

2

Redesign

Redesign

M

8

6

4

2

Redesign

L

10

8

6

4

2

N

10

10

8

6

4

Assign to Inspection Schedule – Step 12

See Step 12.

Include the completed PRV RBI assessment into the scope of an Inspection Schedule that represents relief valve recertification.

Specify activities during recertification, e.g.,

  • Visual
  • As received pop test
  • Disassembly/reassembly & visual
  • Set Pressure
  • Pop test

Was this helpful? Click to add feedback comments

Changing your password will log you out immediately. Use the new password to log back in.
First name must have atleast 2 characters. Numbers and special characters are not allowed.
Last name must have atleast 1 characters. Numbers and special characters are not allowed.
Enter a valid email
Enter a valid password
Your profile has been successfully updated.
ESC

Eddy AI, facilitating knowledge discovery through conversational intelligence