Two of the leading causes of user concern regarding
control-valve performance may surprise you. Field experience
shows that the biggest contributor to control-valve maladies is
often oversizing. It is a classic example of having too many
cooks in the kitchen. Each of the engineers contributing to the
final process specification adds what he or she believes is a
constructive safety factor, and the resulting installed product
is substantially oversized for the application. However, the
focus of this article is what may be the second-leading cause
of concerns: shutoff performance that does not meet the end
In all fluid-processing industries, tight-sealing control
valves are vital to ensuring product quality, efficiency,
safety and environmental protection. But how do you determine
how tight is tight enough?
Tolerance of leakage can vary widely from application to
application; tight enough in one case can be overkill in
another and insufficient in a third. And to top it off, the
various industry standards that classify seat leakage in
industrial valves fail to address some of the practical issues
that confront valve manufacturers, specifiers and end users. In
fact, it is quite possible to successfully specify, manufacture
and test a valve according to a well-established industry
standard, yet still experience less-than-satisfactory results
in the field.
This article will help address the technical and practical
issues related to seat leakage, discussing the fundamentals
behind the governing industry standards and offering guidance
that users can apply to enhance initial seat leakage
performance and help extend the life of their valve assets.
Defining seat leakage.
Seat leakage is defined as leakage that is internal to a
valvebetween the inlet and outlet sides of the
valvewhen the valve is in its closed position. It is not
limited to leakage across the valve seat, but also encompasses
all leakage across the valve trim when the valve is in the
closed position. Leakage across internal trim seals, such as
piston rings, and across trim-to-body seals, such as gaskets,
can be counted as seat leakage.
It is important to note that, while leakage through valve
stem packing is of growing concern in the industry, governing
industry standards address this type of leakage separately and
do not consider it to be a form of seat leakage.
Industry gold standard.
ANSI/FCI 70-2, Control Valve Seat Leakage,
published by the American National Standards Institute and the
Fluid Controls Institute, is widely recognized as the defining
standard for leakage in control valves. It categorizes seat
leakage into six groups (Class I to Class VI). Generally, each
higher class defines tighter or more stringent leakage
criteria. While these classes are not linear in their
progression, they follow a logical succession from lowest to
highest class with some added footnotes.
Additionally, for each leakage class, ANSI/FCI 70-2
provides detailed test procedures and defines the maximum
allowable leakage (MAL). The test procedures include provisions
for using water and/or air as the test medium, along with
allowable pressure and temperature ranges. IEC Standard
60534-4, Industrial Process Control ValvesPart 4,
Inspection and Routine Testing (2006), published by the
International Electrotechnical Commission, contains the same
control valve seat leakage criteria.
Class I leakage is relatively undefined and is simply stated
as agreed upon between purchaser and manufacturer. It does not
identify a test procedure or specify a standard test pressure
and, therefore, does not define the maximum allowable
So why does it exist? While not stated in the standard,
Class I is normally used when the valve specification has no
leakage-specific criteria based on its application. Possible
examples include a valve that is always in the open position,
or a system in which the control valve is always accompanied by
an adjacent remote-operated isolation valve. In these cases,
the user has no expectation that leakage through the valve will
negatively impact the process, so Class I may be
The definitions of the other leakage classes are not quite
as liberal. The Class II, III and IV definitions share many
common attributes and are easily discussed as a set. For each
of these classes, there is a common test procedure that permits
the use of water or air at a temperature of 50°F to
125°F and a pressure of 45 psig to 60 psig. The maximum
allowable leakage for each class is expressed as a function of
the rated valve capacity as follows:
Class II: 0.5% of rated valve capacity
Class III: 0.1% of rated valve capacity
Class IV: 0.01% of rated valve capacity.
There are two observations to note at this stage. First, the
difference between these classes is not trivial. Class III is
five times tighter than Class II and Class IV is 50 times
tighter than Class II.
Second, a value such as 0.01% of rated valve
capacity is not of much use to someone who wants to know
how many cups or buckets or gallons of leakage per minute is
acceptable for a particular valve. The fact is, rated
valve capacity can be a rather elusive value to those who
do not crunch valve data every day.
Fortunately, control-valve manufacturers rate valve
capacities in terms of Cv, the universal
flow coefficient for valves as defined by ANSI/ISA Standard
S75.01.01, and a basic conversion to
Cv can help put the rated valve
capacity leakage criteria into a more tangible
Per ANSI/ISA Standard S75.01.01,
Cv is defined as the number of gallons of
water per min., at 60°F, that will pass through a flow
restriction with a differential pressure of 1 psi. The equation
Cv = Flow coefficient
Q = Flow quantity gallons per min. (gpm)
SG = Specific gravity
dP = Differential pressure psi.
When quantifying seat leakage, the flow restriction is the
subject valve. Assuming that valve capacity
(Cv) is known, the equation can be solved
for flow (Q) as follows:
The SG of water between 50°F and 125°F only varies
from 1.0007 to 0.9887 and thus can be rounded to 1.0.
Therefore, the MAL for Classes II through IV can be expressed
in terms of gpm as a function of test pressure as:
To illustrate how these equations play out in the real
world, consider a common 10-in. nominal, cage-guided,
metal-seated control valve that meets standard ANSI/ISA
75.10.02 face-to-face dimensions. A typical
Cv for this type of valve is 950. Using 50
psi of water at 60°F, the MAL rates for this valve
Class II: 33 gpm
Class III: 6.7 gpm
Class IV: 0.67 gpm.
Class V represents what is commonly referred to as an
effectively zero-leakage control valve. It should
be remembered, however, that the Class V category still retains
some allowance for leakage. And, as will be discussed later, a
seemingly zero-leakage shop test does not always lead to zero
leakage under service pressures and temperatures. Class V,
however, is very tight and certainly takes a step far beyond
A different test procedure is used to calculate maximum
allowable leakage for Class V. Rather than a test pressure of
45 psig to 60 psig, Class V requires the use of water at a
pressure differential that is +/-5% of the service pressure
differential, not to exceed the 100°F-rated pressure of the
valve body. The formula for MAL is:
Class V (ml/min.): (5 x 104) x
(seat diameter in in.) x (test pressure in psi). (6)
For the purposes of comparison, consider the 10-in. valve in
the previous example and assume that the service pressure
equals the 50-psig test pressure. The Class V MAL would be:
Class V: (5 x 104 ml/min.) x (10
in.) x (50 psi) = 0.25 ml/min. = 6.6 x
Roughly speaking, this is 10,000 times less leakage than the
same Class IV valves MAL of 0.67 gpm.
Like Class V, Class VI bears no relationship to the other
classes. The specified test procedure requires the use of air
or nitrogen gas at a temperature of 50°F to 125°F and
at a pressure differential of 50 psi.
The maximum allowable leakage is not given in equation form,
but is instead presented as a table of values per nominal valve
size. And, since the test pressure is fixed, the MAL values are
discrete. The values range from 0.15 ml/min. for a 1-in.
nominal valve to 28.4 ml/min. for a 16-in. nominal valve.
Valves that are 8-in. nominal and smaller are also given an
equivalent MAL in bubbles per minute.
Fig. 1. An unbalanced
valve plug. The solid plug
leaves one potential leakage
path at the seat.
Several aspects of the ANSI standard can be confusing and
frustrating for users. For starters, the maximum allowable
leakage for Classes II, III and IV do not vary with service
pressure. The test pressure for these classes is fixed between
45 psig and 60 psig, while most control valves operate at
pressures above that range. For example, the 100°F maximum
working pressure of ASME B16.34 Class 2500-rated
valves is in the neighborhood of 6,250 psig.
It may be tempting to work around this by always opting for
a Class V valve, but this is not a practical solution. As
illustrated earlier, a Class V valve can be up to 10,000 times
tighter than its Class IV cousin. That is obviously a huge leap
in performanceone that, understandably, comes with a
higher price tag. The frustration comes for a user faced with a
situation in which a Class IV valve is not tight enough but
Class V performance is far more than is required. The ANSI
standard offers no middle ground.
Note that the 2006 revision of ANSI/FCI 70-2
permits the use of air at 50 psig for the Class V test. While
this adds convenience, it negates one of the attractive
attributes of the Class V criteria, which is the use of a test
pressure that matches the service pressure.
Because the difference between Classes IV and V is so large,
it may be prudent for the end user to consider the value of the
process fluid when making a decision on leakage class. Over the
life of a control-valve trim, when leakage means lost revenue
or lost energy, the additional cost of an upgraded trim can pay
for itself in a short period of time.
The parallel standard, IEC Standard 60534-4, does
offer an intermediate step between Class IV and Class V, known
as Class IV-S1. It uses the Class IV test pressure, with MAL
Class IV-S1 MAL = (5 x 106) x rated
Solving for gpm, the equation becomes:
Class IV-S1 MAL (gpm) =
or 100 times less leakage than Class IV.
Returning to the 10-in. example valve at 50 psig test
Class IV-S1 MAL = 0.0067 gpm.
For Classes II through IV, neither ANSI/FCI nor IEC supports
the use of test pressures that approach service pressures.
Class V MAL does at least use a representative (dP)
value, but even it can leave the informed user wanting.
Recall Eq. 2, the driving equation behind the
This is the fundamental equation that the valve industry uses
to engineer, sell and deliver flow capacity. It states that
flow (Q)or, for the purpose of this discussion,
leakageis a function of the inverse of the square root of
differential pressure (dP).
Then consider the equation for MAL for Class V (Eq.
Leakage = (5 x 104 ml/min.) x (seat
diameter) x (dP).
Here, the flow is directly proportional to the pressure
differential. Needless to say, Eqs. 23 produce diverging
If it can be assumed that the Eq. 2 curve reflects the
fundamental relationship between flow and differential
pressure, then even the more sophisticated Class V definition
has to be taken in narrow context. It is no wonder that
ANSI/FCI 70-2 states that the standard cannot be
used as a basis for predicting leakage at conditions other than
those specified. In other words, the standard should not
be used to predict leakage at conditions other than the test
conditions. This leaves the user with little or no prediction
of field performance.
Also note that the 10,000:1 example shown previously only
applies to the test pressure of 50 psi. Due to the disconnect
between the driving equations for MAL and classical physics of
flow through a restriction, comparison of the two equations at
alternate pressures yields alternate ratios.
For instance, in the previous 10-in. example valve, the
ratio at a 1,000 psi test pressure is:
(Class IV) / (Class V) = (3 gpm) / (0.0013 gpm) = 2,308.
This discussion of ANSI/FCI 70-2 provides some
insight into the attributes to consider when looking at another
commonly referenced leakage standardMSS SP-61,
Pressure Testing of Steel Valves.
MSS SP-61. The Manufacturers
Standardization Society of the Valve and Fittings Industry
publishes MSS SP-61. Among other topics, it addresses
valve leakage, although it uses the term Seat Closure Test.
It is important to begin by acknowledging that this
standard, by its own definition, does not apply to control
valves, but instead to valves used in full open and
full closed service. More specifically, it is
intended for use with isolation, stop, and check valves.
Nonetheless, its use has been creeping into control-valve
specifications. Thus, it is important to address its
capabilities and appropriate use.
Section 5 of MSS SP-61 defines test procedures as
well as acceptance criteria of seat closure tests. The test
pressure is specified as 1.1 times the 100°F rating, which,
if applied to an ASME B16.34 control valve, means 1.1
times the cold working pressure.
The standard defines only one class of leakage, although
either liquid or gas can be used as the test fluid. The maximum
allowable leakage is specified as:
For a liquid test: MAL = (10 ml/hr) x (valve NPS) where
NPS is nominal seat size in inches.
There are a number of modifying circumstances named in the
standard that can impact test pressure and maximum allowable
leakage criteria. For instance, there are provisions that, for
certain sizes and classes of valves, permit the use of 80 psi
air rather than 1.1 times the cold working pressure.
Additionally, there are provisions that, for certain types of
valves, permit the allowable leakage to be increased by a
factor of four.
Because this standard specifically applies to on/off valves,
rather than to control valves, it is difficult to interpret the
exceptions in terms that clearly apply to control valves. To
help prevent misunderstanding when these exceptions are
applied, it is recommended that they be specifically identified
and integrated into the product specification.
MSS SP-61 can be readily compared to
ANSI/FCI Class V because both standards consider the
nominal seat size and, in a roundabout way, both standards
define MAL as variable in direct proportion to test pressure.
Interestingly, the two standards converge when the test
pressure is 320 psi. Any pressure above that value and MSS
SP-61 is more stringent, while test pressures below that
value show Class V to be more stringent.
API Standard 598. A full
discussion of seat leakage would not be complete without
recognizing API Standard 598, Valve Inspection and
Testing, published by the American Petroleum Institute.
Like MSS SP-61, API 598 is not intended for control
valves but has begun creeping into the control-valve
specification process, so the test parameters are presented
here for comparison.
API 598 uses test pressures that are 1.1 times the
maximum allowable working pressure at 100°F. The MAL is
generally more stringent than ANSI/FCI 70-2, to the
extreme that it is expressed in drops and bubbles per minute.
Like MSS SP-61, this standard was not written for
control valves and further discussion of the unique
requirements of control-valve performance will reveal why.
So now armed with the factual data regarding the various
standards, it is appropriate to discuss how tighter seat
leakage is obtained with various trim designs. While this
article will not provide an exhaustive exploration of control-valve trim
designs, the primary methods used to address seat leakage will
Whether for rotary or reciprocating valves, the most basic
trim design is a simple metal-to-metal interface between the
valve seat and valve plug. This interface is the primary path
for seat leakage, and its design and control is the single
largest contributor to leakage differentiation.
From a broad viewpoint, it would appear that controlling
this leakage is a simple matter of the appropriate mating of
parts. However, even when mating high-quality machined surfaces
together, microscopic imperfections can allow leakage.
Additionally, in spite of superior plug-to-seat geometry, if
the parts are not self-aligning within the trim assembly, the
surfaces will not contact appropriately and repeatedly across
multiple open and close cycles, and the desired seat tightness
will not be achieved.
For example, it is clear that a simple drilled hole that is
0.12-in. in diameter would be considered a leakage path of
unwanted magnitude. Although its equivalent flow area is only
0.011 square in., when tested to Class V standards at 1,000
psi, this hole would lead to flow of roughly 10 gpm. However,
if the same flow area were evenly distributed around the
circumference of a 10-in. plug-to-seat interface, it would
amount to a gap of only 0.0004-in., or approximately 1/10 the
thickness of a human hair. Class V leakage for this 10-in.
valve at 1,000 psi would only allow 0.001 gpm1/10,000 the
rate of the single-orifice flow. Obviously, there is a need for
more than simple dimensional control of the mating parts.
For this primary seat, the plug-to-seat contact geometry and
alignment are critical. The plug should have some means of self
alignment when approaching the seat in the closed
configuration. This can be accomplished with appropriate
lead-in angles and/or continuous guiding along the stem or
cage. The plug-to-seat contact geometry should result in a
single line of contact around the circumference of the parts.
This creates large unit loading, which is critical to closing
the microscopic irregularities of the mating surfaces.
Compliant material interfaces. One method
used to address the inherent imperfections in the plug-to-seat
interface is to utilize a compliant member at the interface.
This can be accomplished with soft materials, such as PTFE,
that are embedded in either part. The compliant material
conforms to the mating imperfections and provides the desired
closure of micro-gaps. This trim configuration would normally
be used in achieving ANSI/FCI 70-2 Class VI
These compliant materials can be used in applications with
temperatures up to 600°F.
A variation of the truly compliant material interface is to
use a metallic member on one component that is softer than its
mating part. This can be accomplished by mating the softer
grades of 300 series stainless steel to harder stainless or
alloy steels. The theory is that, when mated together under
loading, the harder material actually causes the softer metal
to deform. For this reason, these parts are often overloaded at
assembly to create a coining effect. It is important to note
that this coining results in a unique matched pair assembly,
and subsequent substitution of parts, and even disassembly and
reassembly, negates the matched affect.
Key specification considerations.
A fundamental element of all plug-to-seat interfaces is that
part loading between the two components can greatly affect
leakage performance. This loading effect is applied by the
force of an actuator, which is always present in control-valve
applications. During the specification process, the valve
manufacturer should provide guidance (based on an understanding
of the behavior of the assembled parts and their tendency to
align, mate and conform) to ensure appropriate plug-to-seat
Because the various components must work together to deliver
the required loading, particular care must be taken when
pairing a control valve and actuator from different vendors.
The separate parties, or a third-party consolidator, may not
coordinate the offerings. The end user must, therefore,
understand the required load and ensure that it is provided by
the actuation platform and supply pressure.
And because actuation load is a function of not only
actuator type and size, but also of supply pressure, it is
important that the party responsible for packaging the valve
and actuator factor in the entire range of possible supply
loadsnormal, minimum and maximumas they can
positively or negatively impact seating performance. The
obvious misstep would be a supply pressure that is too low to
provide the necessary part loading under service conditions. It
is equally important, however, to ensure that the actuator does
not overload the parts and deform the trim or body components
or cause the plug and stem to engage too tightly and become
wedged together, negatively impacting the control valves
ability to respond to an opening signal.
An additional feature of these simple plug-to-seat
interfaces is that the direction of flow, and thus fluid
pressure, can either assist or detract from part loading. The
common terms used to determine pressure and flow tendencies
relative to trim design are flow-to-open (FTO) and
FTO trim designs. In FTO trim designs, the
flowing fluid and its associated static and dynamic pressures
tend to force the valve plug off of the seat, thus compromising
seat load and increasing leakage tendency. For this type of
trim, it is important that the worst-case conditions are
factored into the design so that fluid pressures do not relieve
the necessary minimum seating load that closes those
micro-gaps. This is especially important in applications
involving metal-to-metal seat interfaces.
FTC trim designs. Alternately, FTC trim
designs are assisted by the fluid pressures, potentially
reducing leakage tendency. Again, there is opportunity for
misapplication when mating valves and actuators. On one hand,
the actuator must be adequately sized to overcome the
worst-case active fluid pressure loads. However, if oversized,
the combination of actuator loading and fluid pressure can
overload a plug-to-seat interface and cause damage, manifested
in excessive part friction and sticking.
It is reasonable to question the use of FTO trim designs
versus their FTC cousins, when FTC would appear to provide the
most generous seat loading and, thus, more effectively reduce
leakage. While this argument is valid, FTO trim designs provide
other beneficial performance traits. They are inherently more
stable, as the flow under the plug does not create the
recirculating flow eddies that can cause the valve to be sucked
into the closed position, a concept known as negative gradient.
Additionally, an FTO design may be preferable based on the
desired failure mode when actuation energy is lost.
In many trim designs, the seating surfaces of both the valve
seat and the valve plug are in the area of flow modulation. In
fact, when a valve plug is lifted just off of its seat, the
inherently small available flow area leads to high fluid
velocity, creating prime conditions for deterioration of
component geometry. Many designs attempt to address this issue
by providing geometry that shields the critical surfaces from
the erosive effects of high-velocity flow. Proper materials
selection is also critical to ensuring that these surfaces
survive the rigors of modulating flow. One of the most
straightforward means to enhance service life in these areas is
to provide high-hardness materials via base materials, process
hardening or weld overlays using hardened alloys.
Collectively, the combination of plug-to-seat interface
geometry, materials selection, actuation loads and pressure
assist forces all come together to create an effective primary
seating surface. The requirements for control valves and on/off
valves are quite different, however. Control valves must be
able to move off of their seated position with a subtle change
in instrument signal. If the above combination creates friction
that is difficult for an actuator to overcome, or if the
friction changes over time, then the control valve is rendered
ineffective in a manner that would be less problematic for an
Balanced trim designs. As noted previously,
these brute force, or unbalanced, trim designs often require
large actuation forces, and thus costly actuation packages.
Early solutions included the advent of double seated valves
(Fig. 2). In these designs, the valve seat geometry is arranged
such that high-pressure fluid is present on both sides of the
valve plug, providing the necessary force balancing to minimize
actuation loads. These designs still exist but their use is
largely limited to smaller nominal sizes.
Fig. 2. In double
valves, the valve seat
geometry is arranged such
that high-pressure fluid is
present on both sides of the
The challenge with double seated designs is that it is
difficult to exactly match the dimensional values between the
two valve body seat regions and the two plug seating regions.
Even when good dimensional matching is provided, thermal
excursions during service can cause differential thermal growth
and loss of seat contact.
Control-valve manufacturers have addressed this dimensional
challenge with the use of balanced trim designs (Fig. 3).
Conventional balanced trims are available in cage-guided
packages, in which the plug travels inside a ported cylinder.
These designs still depend on an effective plug-to-seat
interface for their primary sealing interface, but they also
have passages that provide flow, and thus pressure,
communication between the top and bottom of the plug. This
permits pressure to equalize on both sides of the plug, and
actuation forces are substantially reduced.
Fig. 3. Conventional
plugs have passages that
communication between the
top and bottom of the plug to
minimize actuation forces.
A complicating factor in balanced designs is that a
circumferential seal is required between the plug and cage to
prevent flow from passing through the balancing holes, passing
between the plug-to-cage clearance and then exiting the trim.
These seals are known as the secondary leakage path, and it is
here that the various designs show differentiation. The seal
can be either plug-mounted and work against the stationary
cage, or cage-mounted and work against the modulating plug.
They can be compliant materials, such as PTFE or flexible
graphite, or metal seals, such as classic piston rings.
PTFE seals. The design range of PTFE has
recently been extended into the 600°F realm and these seals
show good resistance to the expected frictional wear of a
modulating seal. PTFE can also be formed to create
pressure-assisting lips that improve the contact between the
stationary and modulating parts. Above the 600°F
temperature threshold, however, PTFE is not viable and flexible
graphite is the only widely available compliant option.
Flexible graphite seals. Like PTFE,
flexible graphite provides the necessary compliance to
dynamically deform to fit its mating parts. The primary issue
with graphite is that the normal frictional wear of a
modulating control valve quickly leads to loss of mass, and
thus volume, in the graphite seal, causing it to lose its
radial contacting load.
Unlike graphite stem packing, internal trim seals cannot be
adjusted by tightening a packing bolt or using a live loading
mechanism. Instead, the frictional wear leads to increased
leakage at the secondary plug seal, and even though the primary
plug-to-seat geometry remains effective, the user experiences
Metallic piston ring seals. For
high-temperature applications, these secondary seals are more
typically metallic. Conventional piston ring designs are widely
used as trim seals for balanced designs. A key differentiator
among various piston rings is the design of the gaps where the
open ends meet. More advanced piston ring designs often use an
inner expander ring to create continuous radial loading of the
outer ring. Good piston ring design leans on installed
circularity, face sealing with the mating groove, and an
ability to install the ring without deforming it.
Note also that for cage-guided valves, these secondary
piston ring seals also provide the dynamic stability that is
necessary for valve plugs with balancing holes. These rings
prevent continuous flow across the balancing holes, which can
cause pressure pulsations between the top and bottom of the
plug and compromise valve stability. This is one of the many
areas where control-valve trim has additional dynamic
requirements that are not required in on/off valves.
Some recent innovations in secondary sealing technology attempt to combine many
of the previously discussed attributes into a single advanced
seal. These include metallic seals that are also compliant to
provide extended wear. Some compliant metals seals have been
utilized to remedy the problems associated with double seated
valves by providing dimensional forgiveness between the two
seating regions. Some designs also have used separate dynamic
seals along with static seating seals to keep the wear of
modulation isolated from the static seal.
Each of these secondary sealing techniques has merit, but
each also has its Achilles Heel in such areas as durability,
scalable size, or temperature gradients. The user should seek
evidence of long-term application experience in installations
that are similar to the proposed application.
Shop-tested vs. installed leakage. A final
topic that must be addressed in this discussion of secondary
seals is the issue of shop-tested leakage versus installed
leakage. A graphite seal in a new assembly, for example, will
typically pass a shop test with ease, even when designed to
meet ANSI/FCI 70-2 Class V or MSS SP-61
criteria. However, a short time in modulating service can lead
to loss of graphite mass and compromised leakage performance.
The fact remains that none of the seat leakage standards makes
any guarantees regarding in-service leakage. They simply
represent stand-alone shop tests of new components, and thus
can be misleading if the user attempts to equate them to field
A popular option for high-temperature applications that
cannot use nonmetallic compliant seals is the pilot-balanced
plug (Fig. 4). This design is a hybrid of unbalanced and
balanced designs, incorporating the best attributes of
designs have a small
unbalanced plug that is
integrated into the larger
primary plug to create a
hybrid balanced design.
In the pilot-balanced concept, a small unbalanced plug is
integrated into the larger primary plug. When the pilot plug is
opened, it provides a balancing path for pressure equalization
across the primary plug. When both the pilot plug and primary
plug are closed, the assembly is effectively unbalanced.
Pilot-balanced plugs are time-tested and in widespread use.
The greatest challenge in working with them is ensuring that
the proper stiffness is maintained during mid-travel
modulation, as inadequate stiffness can lead to instability.
Most designs use internal springs, either a coil or Bellville
type, to maintain a target stiffness between the pilot plug and
the primary plug. Others incorporate pressure ports to utilize
the adjacent high-pressure fluid in a chamber with a pressure
area biased toward stiffness.
The positioners role.
In addition to the mechanical variations that can impact
seat leakage performance, the brain on top of the
control valvethe positionercan be an important tool
in protecting critical seating surfaces. As mentioned earlier,
valve trim that operates just off of its seat creates small
flow passages and the resulting high velocities within the trim
are highly erosive, even in clean fluid
applications. Many of todays digital positioners have
options for an override that prevents the plug from operating
in this unfavorable travel position. These overrides are
typically user-configurable, allowing the user to determine the
desired minimum travel position relative to both trim
protection and required low-end capacity.
Additionally, the most advanced digital positioners can
remember their shop-tested attribute of stem position when the
trim is seated. This is quite useful if foreign material or
damaged trim components prevent the plug from properly seating
on the valve seat. Even though the actuation force may suggest
that the plug has reached its proper travel stop, these smart
positioners know that this position does not signify true
seating and can provide the necessary alarm to process control
Clearly, there are a multitude of factors that influence
seat leakage measurement and performance in control valves.
While there are various industry standards that can be utilized
in the specification process, ANSI/FCI 70-2 is the
most widely used despite its shortcomings. Given that a
shop-based seat leakage test provides limited prediction of
field performance, the end user is wise to understand the more
durable attributes that lead to satisfactory seat leakage
performance over the installed life of a control valve.
Don Sanders is product manager for
engineered products and the severe service segment for
GE Energy, www.ge-energy.com. A 29-year veteran of the
control-valve industry, he has held positions in
engineering, manufacturing, marketing and management.
He has a BS degree in mechanical engineering from the
Georgia Institute of Technology.