Compression packing is found in applications ranging from transmission of natural gas and water to caustics and high-temperature steam. When used properly, it is a cost-effective, high-performance means of sealing. Unfortunately, compression packing creates friction, which can cause major issues in certain applications. Knowing how to reduce that friction can be critical in minimizing those issues.
WHAT’S USED AND WHY
Users of air- and motor-operated control valves (AOV and MOV) generally desire low-friction packing that allows accurate, efficient and consistent actuation while maintaining an effective seal on media. The frictional force packing exerts on a dynamic surface is mainly a function of the type of material, contact surface area and compressive load. Although other system variables and inputs also affect friction, they are more difficult to quantify or modify.
Friction reduction strategies involve modifying packing materials, configurations and installation procedures to attain target frictional loads. Different applications have different allowable leakage rates. Effectively sealing one application may call for graphite, while another may require a PTFE-based packing to reduce friction. Users also may have preferences based on cost, logistics or historical performance. There is no single solution to all sealing issues; the strategies discussed here conceptually apply to most applications, but they need to be validated before implementation. Each application will have an optimal solution that takes into account currently available sealing technologies and strategies.
Compression packing serves as a barrier to slow media migration from the higher-pressure system to the lower-pressure surrounding environment where the valve is operating. The sealing mechanism of compression packing is based on a tight fit between the packing and dynamic sealing surface. This fit is created by applying axial compression that causes radial movement of the packing against the sealing surface1. Figure 1 illustrates the dynamics of axial compression and radial expansion.
HOW ESCAPES OCCUR
The magnitude of a leak is determined by system variables such as media, shaft finish, pressure, braid construction and installation, shaft run-out and temperature. An important point here is that friction and sealing are separate, but related issues with regard to compression packing. Typically, optimized solutions are a combination of both of these factors. Hypothetically, one could achieve low friction by not installing any sealing at all; but the result would be a leaking valve. Conversely, excellent sealing could be achieved by welding a stem to the valve bonnet; but the valve could not be actuated. Realistically, then, the factors within an operator’s immediate control for emissions include the type and number of packing rings used, proper installation and axial loading.
Friction is not the major consideration in most sealing applications. Generally, fugitive emissions, chemical compatibility and continuous temperature capability are the main criteria for selecting one packing solution or installation strategy over another.
There are several reasons this is true. For example, in the petrochemical industry, fugitive emissions are of greater concern than friction in isolating valves. In the solar industry, solar companies are operating valves at 1,000°F (538°C) in molten salt, so they are more interested in temperature and chemical resistance.
In applications involving AOV and MOV, the size of the actuator required to effectively move the stem needs to be determined. Correctly sizing the actuator requires knowing the frictional force exerted by the seal as well as the force needed to control the media stream. Control valves require consistent friction to accurately modulate process flows.
Another factor that must be considered is “stiction,” a term used to describe an actuating stem catching periodically on the packing set and causing the stem to move erratically. This leads to issues in effectively modulating flow. Figure 2 depicts a sample packing set that does not exert consistent friction on a stem versus a set that achieves that consistency.
There are three basic strategies for reducing friction while maintaining an effectively sealed system. These include reducing the load on the packing set, reducing the number of rings and changing the packing material. A simplified formula developed by Electric Power Research Institute (EPRI)2 for calculating friction on a reciprocating shaft is:
Where F is friction force (pound-foot or lbf); D is stem diameter (inch or in.); H is uncompressed packing height (in.); s is compressive stress (psi); µ is friction factor; and Y is axial/radial stress factor (commonly 0.5).
Reducing the axial compressive stress (s) reduces the radial stress exerted by the packing set on the stem. Figure 3 depicts the same packing set tested at four different loads. The relationship between compressive stress and friction is shown to be about linear in Figure 2. An increase in gland stress is about proportional to an increase in friction as the shaft reciprocates.
For a variety of reasons, the systems for a wide range of applications are over-tightened during installation. Most packing sets have a target stress range for effective performance. Beyond that range, extrusion may occur. If compression packing is effectively sealing the application, applying additional stress serves no purpose and can adversely affect performance. The manufacturer or consulting with the maintenance team can provide the target stress for effective performance. Assuming well lubricated alloy steel bolts are used, Equations 2 and 3 are commonly used to determine loading conditions.
F = Area * Target Compressive Stress
Where F is bolt force (lbf); A is gland sealing area (square-in. or sq. in.); and s is target stress on compression packing (lbf/sq.-in).
Where T is required bolt torque
(in.-lbf); F is bolt force (lbf); w is nominal bolt diameter (in.); and B is number of gland bolts.
Reducing the number of rings in a packing set limits the contact area with the shaft. In Equation 1, this lowers the uncompressed packing height (H), which is proportional to a lower frictional force. In theory, most of the applied stress impacts only the two rings closest to the gland follower. These rings provide most of the sealing effectiveness; the remaining rings provide little sealing, but add to the overall frictional force exerted on the moving shaft.
Figure 4 charts actual test data correlating the number of packing rings with required actuation force. Adding rings increases friction, but this is not a linear relationship, and it varies significantly from braid to braid depending on the material and construction. Removing rings can pose potential issues with spacing and sealing effectiveness. Spacing can be maintained by installing carbon or steel bushings that maintain the height of the packing set without contacting the shaft. The number of rings required for a seal depends on the application and should be determined by those familiar with the system.
Changing the packing material to one with a lower coefficient of friction (COF) reduces friction. In Equation 1, the friction factor (µ) quantifies how the packing material resists movement on the dynamic sealing surface. A friction factor is not the same as a COF, however. Friction factors are lumped variables describing friction for a specific configuration or braid. This is different than the COF, which describes an inherent material property. Friction factors vary for different types of compression packing.
For example, a PTFE-based braid may have a published value of 0.08; a graphite braid with lubrication might be about 0.09; and a die-formed graphite set may be near 0.1. These friction factors vary from actual values because of manufacturers’ safety factors, consideration of worst-case scenarios, and averages over different sizes and styles of braids. Figure 5 shows the actuation force required for a four-ring set of 3/8-in. PTFE-coated carbon fibers, a die-formed pure graphite set, pure PTFE fiber with a lubricant dip and PTFE fiber braid with lattice construction.
MATERIALS OF CONSTRUCTION
Graphite and PTFE are the predominant low-friction materials for compression packing. PTFE is a highly lubricious material, but is limited by its 500°F (260°C) temperature rating, as well as by high creep and flow characteristics. Graphite can withstand temperatures of up to 850°F (454°C) in oxidizing atmospheres and 1,200°F (649°C) in steam atmospheres. Both these materials can be used as the dominant material of construction or can be added to reduce friction. Graphite, PTFE and other polymers and lubricants are commonly added through a dip or dispersion to reduce friction during operation. They also can be manufactured into pure PTFE or graphite sealing products.
Typically, graphite is formed into a sealing product by die-forming flexible graphite foils into solid rings. PTFE can be formed into fibers and braided, similar to other fiber braids. PTFE and graphite materials also can be processed with other fibers and fillers to optimize desired characteristics such as lower friction and resistance to extrusion. For example, a thin coating of PTFE on carbon or graphite braid can significantly reduce friction, while the carbon core maintains the structural integrity and creep resistance of the braid.
Another sealing solution is to use die-formed graphite sets with angular planes that promote radial movement, which minimizes the compressive load required to seal effectively. This reduction in compressive load coupled with graphite’s soft material properties creates an effective seal and a reduced frictional load on the traveling stem. Rather than applying a high frictional force, soft graphite rings will deform to a balance point between the shearing frictional force and the material strength. In addition, the reduced compressive load required for sealing means the end rings, typically a more robust braided material, see less compressive load and subsequently apply a lower frictional force on the traveling stem. Referring to Equation 1, this generally means less frictional force is generated using a die-formed set in comparison to an equivalent braided material. PTFE over carbon exhibited the lowest friction of the braids tested in Figure 5.
There are no standard test methods for friction induced by compression packing, so manufacturers develop their own standardized tests for comparing the frictional properties of different products. Standard tests typically measure COF. ASTM G1115-103 provides a guide for measuring and reporting COF under specific, controlled settings. However, the results of analyzing a specific portion of braid for friction are not particularly useful since stem friction in a valve results from the dynamic interaction of constantly changing variables such as lubrication, finish, temperature and cycle number. (See “Testing Protocol” to read the test protocol Garlock developed using friction factors and simulating actual applications to compare the field of products.)
There are other factors that directly impact the friction generated by a compression packing set, but they are more difficult to measure and control in the field than a simple change of packing material. Among them is shaft finish, which is recommended at 32 micro inches (AARH) or better on a reciprocating valve stem. Run-out of a stem or traveling shaft unevenly loads and unloads the packing, potentially exceeding the limits of the material’s compressibility and recovery properties, which detrimentally affect sealing. In addition, the gland follower may interfere with the traveling stem.
Improper installation is the most common cause of sealing failure, so proper installation is of critical importance for compression packing. Installation issues can adversely affect applications such as AOV and MOV control valves, which require low, consistent frictional force for efficient performance and lower long-term operating costs. Typically, there is no torque issue with MOV units, but friction impacts accuracy and power usage. The distributor or manufacturer of a particular sealing product should be consulted for best practices in installation.
The required actuation force on a valve’s actuating mechanism depends on several factors. Selection of the best sealing solution for a particular application must take into account the key parameters that affect that force, most notably size, temperature, application, media, pressure and speed (STAMPS).
- Size. There are standard sizes for many components: for example, API valve stems. Non-standard sizes should be conveyed to sealing manufacturers in the form of dimensional drawings. Some applications may require field measurements.
- Temperature. A major consideration is the continuous temperature to which the packing will be exposed, including high/low excursions as well as any regular thermal cycling inherent in the process. Note that the frictional heat generated by rotating equipment will increase the temperature of the fluid contacting the seal. Typically, movement does not generate excessive heat in valves. Temperature data will immediately limit the number of viable seals for an application.
- Application. Defining the parameters of a particular application requires information about where the seal will be installed. Selection of valve compression packing will depend upon the condition of the stem, temperature and chemical requirements, whether its motion is reciprocating, helical or continuous, and whether a specific level of leakage must be attained to meet regulatory requirements.
- Media. Common chemical nomenclature or trade names are used to identify the media that will come into contact with the seal. Some processes employ secondary media as well. For example, a food-processing line that is flushed once a day with a sodium hydroxide solution calls for a seal that is compatible with both this corrosive medium and the food being processed.
- Pressure. This refers to the internal pressure a seal must contain. Most systems operate at fairly consistent pressures, but as with temperature, it is important to know if the seal will be subject to pulses and other variations as a normal part of operation.
- Speed. The speed of a rotating shaft or reciprocating rod must be taken into account when selecting compression packing for dynamic applications. High speeds call for sealing materials that can withstand and effectively dissipate frictional heat. Speed is typically not a factor for valve applications.
- Armed with this data, the best combination of product and sealing strategy can be determined for minimizing friction and more effectively operating control valves.*
*AUTHOR NOTE: A good paper that summarizes some of the points made here is “How stem finish affects friction and fugitive emissions with graphite-based control valve packing,” by James Walker, presented at Valve World 2010. http:// www.jameswalker.biz/en/pages/100-technical-papers
1Fluid Sealing Association. What is the Impact of Packing Friction on Equipment Performance? February 2009. Web. www.fluidsealing.com/sealingsense/ Feb09.pdf.
2EPRI: Electric Power Research Institute. “6.4.2 Initial Friction Results.” EPRI - NP 5697. 1988. Print.
3ASTM International. “Standard Guide for Measuring and Reporting Friction Coefficients1.” G115-10. 2010. Print.