Chemical Engineering July 2010 - 47

Relief System Deficiency
Undersized
relief device
Improperly
installed
relief device
TABLE 1. RELIEF-SYSTEM PROBLEM IDENTIFICATION DURING OVERPRESSURE-SCENARIO MODELING
Identified When:
Insufficient reliefdevice
area
Excessive relief-valve
inlet-line pressure drop
Excessive relief valve
backpressure
Incorrect relief-valve
set pressure
Calculated relief-device area > installed relief device area
Friction pressure drop in pressure-relief-device inlet line > allowable friction
losses (typically 3% of the set pressure)
Relief valve backpressure > allowable backpressure (typically 10% for conventional
valves, or 50% for balanced bellows valves considering backpressure
capacity-reduction factor)
Pressure in protected vessel or line > Maximum allowable accumulated pressure
(typically 10, 16 or 21% of the MAWP for pressurized vessels with single
relief valves for non-fire scenario, multiple relief valves for non-fire scenario, and
relief valves for fire-scenario), AND pressure at PSV inlet < PSV set pressure
Miscellaneous Excessive line velocity Line Mach number > allowable Mach number (typically 0.7)
Insufficient knockout
drum liquid separation
Excessive flare radiation
Effectively
separated droplet size at maximum relief load > allowable droplet
size (typically 300-600 µm)
Calculated radiation level at a specific point > allowable radiation level (typically
1,500 Btu/h-ft2 where presence of personnel with adequate clothing is
expected for 2-3 min during emergency operations, or 500 Btu/h-ft2, where
continuous presence of personnel is expected, both including sun radiation)
refers to an installed relief device
with insufficient capacity to handle
the required relief load. The third
type encompasses relief devices with
incorrect set pressures, possibility
of involuntary blocking or hydraulic
problems. In addition to these problems,
other less frequent ones can be
cataloged as miscellaneous deficiencies.
A relief-system problem tree is
shown in Figure 1.
In a previous statistical analysis of
272 process units in the U.S., it was
observed that [2]:
* 15.1% of the facilities lacked relief devices
on equipment with one or more
potential overpressure scenarios
* 8.6% of the relief devices were undersized
*
22% of the relief devices were improperly
installed
Identifying potential problems
There are work methodologies that
allow identifying potential problems
in relief systems. OSHA regulation
29 CFR 1910.119 is based on safety
audits that use techniques such as
process hazard analyses performed at
regular intervals. The work methodology
established by this regulation to
identify safety hazards comprises two
basic steps [3]:
1. Process safety data gathering,
which includes the following:
* Process chemical safety data
* Process technology data
* Process equipment data [materials
of construction (MOCs), piping and
instrumentation diagrams (P&IDs),
design standards and codes, design
and basis of design of the relief systems,
among others]. As part of these
data, " the employer shall document
that equipment complies with recognized
and generally accepted good
engineering practices " [3]
2. Process hazards analysis, which
may include: What-if, hazard and operability
(HAZOP) study, failure mode
and effects analysis (FMEA), fault-tree
analysis or equivalent methodologies.
In order to document that the plant
equipment complies with recognized
and generally accepted good engineering
practices, the plant management
must validate that the facilities are
protected against potential overpressure
scenarios, in accordance with
accepted codes and standards, such
as API standards 520 and 521. An effective
relief-system-validation study
comprises the following steps:
1. Plant documents and drawings
gathering. The first step involves
obtaining and classifying the
existing plant documents and drawings:
process flow diagrams (PFDs),
mass and energy balances, product
compositions, equipment and instrument
datasheets, P&IDs, relief device
datasheets, relief loads summaries, relief
line isometrics, one-line diagrams,
unit plot plan, and so on.
2. Plant survey. The second step consists
of inspecting the installed relief
devices to verify that they are free of
mechanical problems, to update and
fill-out missing data in the plant documents
and to verify consistency between
the documents and drawings
and the actual as-built plant. During
plant surveys, other typical indications
of relief system problems are the
presence of pockets, leaks or freezing
in relief lines and headers.
3. Overpressure scenario identification.
In this step, the P&IDs are
examined in order to identify credible
overpressure scenarios for each piece
of equipment.
4. Overpressure scenario modeling.
The fourth step is to model each
credible overpressure scenario. Each
model is developed in accordance with
the chosen reference standard (for
instance, API 520 and 521). The following
calculations are typically performed
during this step:
* Required relief load for each overpressure
scenario
* Required relief-device orifice area
for each overpressure scenario
* Relief line's hydraulics
* Knockout drum (KOD) liquid-separation
verification
* Flare or vent radiation, dispersion
and noise level calculations
The overpressure scenario modeling
can be done in different ways, be it
by hand calculations, spreadsheets
or by the use of steady-state or dynamic
relief-system simulation software.
The results of the models are
analyzed to identify potential problems.
Table 1 summarizes the possible
relief system problems and the
ways to identify them on the calculation
results.
Available solutions
There are various solutions for each
type of relief system problem. The
available solutions can be classified
as: (a) modification of existing relief
system components, (b) replacement
of existing relief system components,
(c) installation of new relief system
components, or (d) increasing the reliCHEMICAL
ENGINEERING WWW.CHE.COM JULY 2010 39

http://WWW.CHE.COM

Chemical Engineering July 2010

Table of Contents for the Digital Edition of Chemical Engineering July 2010