As published in the February and March 2007 issues of Chemical Engineering Magazine

Piping Design Part 1: The Basics

With regard to material of construction, the ongoing evolution of technology has raised expectations throughout industry
William M. (Bill) Huitt W. M. Huitt Co. This is the first in a series of three articles that will cover a wide range of piping topics. Topics that will cross industry lines to include chemical, petroleum refining, pharmaceutical, and other industries as well. It will be the intent of these articles to address questions and misunderstandings as they relate to industry on a general basis. Pipe is pipe. This is a euphemism (jargon if you will) quite often used among piping designers and engineers. Taken at face value, this is a true statementpipe is certainly pipe. However, taken in context, it means that no matter which industry you work in when designing piping systems its all the same. And in that context it could not be further from the truth. The pharmaceutical industry, in its current state of growth, is a relative new comer to design, engineering and construction compared to the oil refining, bulk chemical, pulp & paper and nuclear industries. As a frame of reference the American Society of Mechanical Engineers (ASME) was established in 1880; the American Petroleum Institute (API) was established in 1919; 3-A Standards (for food & dairy) were first developed in the 1920s; the ASME committee for BPVC (Boiler Pressure Vessel Code) Section III for nuclear power was proposed in 1963; Semiconductor Equipment and Materials Institute (SEMI) was established in 1973; the International Society of Pharmaceutical Engineers (ISPE) was established in 1980; and ASME Biopharmaceutical Equipment (BPE) issued its first Standard in 1997. Prior to ASME-BPE much of the 3-A piping Standards were plagiarized to facilitate design of pharmaceutical facilities. While some of the above Standards Committees, and their resulting Codes and Standards, are specific to a particular industry others are more generalized in their use and are utilized across the various industries. As an example, Not only does the design and construction of a large pharmaceutical facility require the need for pharmaceutical based Standards, Codes, Guidelines and Industry Practices such as those generated by ISPE and ASME-BPE, it also requires those Standards created for other industries as well. Meaning that, when designing and constructing a bulk pharmaceutical finishing facility, or a bulk Active Pharmaceutical Ingredient (API) facility the engineers and constructors will also be working under some of the same standards and guidelines as they would when designing and building in other industries such as a petroleum refinery or bulk chemical facility. It is not that the pharmaceutical industry itself is young, but the necessary engineering standards and practices are. Within the past fifteen or so years, industry practice, including dimensional standards for high purity fittings, were left to the resources of the pharmaceutical Owner or their engineering firm (engineer of record). The same applies to construction methods and procedures, including materials of construction. These requirements were basically established for each project and were very dependent upon

what the Owners personnel and the engineering firm brought to the table. Industry standards did not exist. With regard to material of construction, the ongoing evolution of technology has raised expectations throughout industry, but even more so in the pharmaceutical, biopharmaceutical and semiconductor industries. For instance, out of the research and development that went into the Hubble Space Telescope came new methodology and technology to better measure and define the all too tangible limits of surface roughness required in material used in hygienic fluid service contact piping. This is of particular interest to the pharmaceutical, biopharmaceutical and semiconductor industries, where cross-contamination at the molecular level cannot be tolerated in many cases. This requires surfaces to be very cleanable. Surface roughness used to be expressed as polish numbers (ie, #4 or #7) then grit numbers such as 150, 180 or 240). The problem with either of these two methods lay in their subjectivity and their generality. These indicators were not specific enough and the accept/reject result relied too much on a subjective visual verification. There will be more on surface finish requirements in Part II. With acute awareness of the ongoing problems currently faced in the pharmaceutical industry and, for altogether different reasons, the semiconductor industry, various Standards organizations have taken steps to alleviate the consistent problems that have plagued the industry in the past with high purity welding issues, standardization of fittings, and guidelines for industry practice. We will discuss some of the finer points of these issues and in some cases what these Standards organizations, are doing to promote and consolidate some of the better thinking in this industry and in this field. In these early paragraphs it seems as though I am singling out the pharmaceutical industry as the focal point of these discussions. As you will see this is not true. And in saying pharmaceutical I do mean to include biopharmaceutical (biopharm) as well. In making an example of the pharmaceutical industry it is simply an attempt on my part to utilize its relative newness in the development of its own particular brand of standards to give the reader a sense of standards development and how these standards evolve. This article and the two that follow will address metallic piping topics including a discussion on hygienic piping. While non-metallic piping is worthy of discussion it is too broad a topic to try and capture here and will not be a part of these articles. Some of the points that will be covered in this and the following articles are topics such as:

1. 2. 3. 4. 5.

6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

ASME flange ratings, is it 150 and 300 pound flange or is it Class 150 and Class 300 flange? Does the 150, 300, etc. actually mean anything or is it simply an identifier? In forged fittings, is it 2000 pound and 3000 pound, or is it Class 2000 and Class 3000? How do you determine which Class of forged fitting to select for your specification? Corrosion allowance in piping; how do you determine and then assign corrosion allowance? How do you select the proper bolts and gaskets for a service? How is pipe wall thickness determined? What is MAWP? What is Operating and Design Pressure? What is Operating and Design Temperature? How do Design Pressure and Temperature relate to a PSV set point and leak testing? What Code should you be designing under? What kind of problems can you expect with sanitary clamp fittings? How do you alleviate those problems with sanitary clamp fittings? What is ASME-BPE? How does ASME B31.3 and ASME-BPE work in concert with one another? What is ASME BPE doing to bring accreditation to the pharmaceutical Industry? Design is the culmination and application of industry standards and industry requirements that take into account constructability along with maintenance and operational needs. These points will be covered as well.

We will first of all lay some groundwork by beginning with the basics of general piping. By understanding the basic elements of piping the designer and engineer can improve their decision making in the material selection process and system design effort. These articles will also make clear a number of misconceptions with regard to terminology and general practices. What we will try to avoid is a lot of in-depth discussion and elaborate analysis on specific points. What I would like to achieve is a general discussion on many topics rather than finite rhetoric on only a few. With that said, this first article is entitled:

Piping Design Part I The Basics

This article will not attempt to cover all of the various types of piping components and joints that are available in industry today. To keep the discussion focused we will discuss only that segment of joints, fittings and components most frequently used in general piping design.

Neither will we duplicate the dialog of previous articles that have provided excellent commentary on segments of this same topic. Articles such as the one written by John C. Cox and published by Chemical Engineering for their January 2005 edition titled Avoid Leakage in Pipe Systems. John provides a concise and descriptive narrative on threaded and compression type connections. And the article by Trinath Sahoo published by Chemical Engineering for their June 2005 edition titled Gaskets: The Weakest Link. In his article Trinath gives the reader some excellent insight into the mechanics of gasket selection and design. PIPE FLANGES Pipe flanges are used to mechanically connect pipe sections to other pipe sections, inline components, and equipment. Flanges also allow pipe to be assembled and disassembled without cutting or welding, eliminating the need to issue a burn card for cutting and welding when dismantling is required. In providing a breakable joint, flanges unfortunately provide a potential leak path for the service fluid contained in the pipe. Because of this, as in all other joints, they need to be minimized where possible. The most prevalent flange standards to be used in industry are based on requirements of the American Society of Mechanical Engineers (ASME) Standards. These include: B16.1 Cast Iron Pipe Flanges and Flanged Fittings, B16.5 - Pipe Flanges and Flanged Fittings (NPS 1/2 through NPS 24), B16.24 Cast Copper Alloy Pipe Flanges and Flanged Fittings, B16.36 Orifice Flanges, B16.42 Ductile Iron Pipe Flanges and Flanged Fittings, Large Diameter Steel Flanges (NPS 26 through NPS 60) B16.47 Large Diameter steel flanges (NPS 26 through NPS 60)
NPS, indicated above, is an acronym for Nominal Pipe Size.

Figure 1 The Threaded flange, through Class 400, is connected to threaded pipe in which the pipe thread conforms to ASME B1.20.1. For threaded flanges in Class 600 and higher the length through the hub of the flange exceeds the limitations of ASME B1.20.1. ASME B16.5 requires that when using threaded flanges in Class 600 or higher Schedule 80 or heavier pipe wall thickness be used, and that the end of the pipe be reasonably close to the mating surface of the flange. Note that the term reasonably close is taken, in context, from Annex A of ASME B16.5, it is not quantified. In order to achieve this reasonably close requirement the length of the thread has to be longer and the diameters of the smaller threads become smaller than that indicated in ASME B1.20.1. When installing Threaded flanges Class 600 and higher, ASME B16.5 recommends using power equipment to obtain the proper engagement. Simply using arm strength with a hand wrench is not recommended. The primary benefit of threaded flanges is in eliminating the need for welding. In this regard Threaded flanges are sometimes used in high-pressure service in which the operating temperature is ambient. They are not suitable where high temperatures, cyclic conditions or bending stresses can be potential problems. Socketweld Flange

Flanges are available with various contact facings (the flange-to-flange contact surface) and methods of connecting to the pipe itself. The flanges under B16.5 are available in a variety of styles and pressure classifications. The different styles, or types, are denoted by the way each connects to the pipe itself and/or the type of face. The type of pipe-to-flange connections consist of Threaded, Socket Welding (or Socket Weld), Slip-On Welding (or Slip-On), Lapped (or Lap Joint), Welding Neck (or Weld Neck), and Blind. Figure 2 Threaded Flange The Socketweld flange is made so that the pipe is inserted into the socket of the flange until it hits the shoulder of the socket. The Pipe is then backed away from the shoulder approximately 1/16 before being welded to the flange hub. If the pipe were resting against the shoulder (This is the flat shelf area depicted in Fig. 2 as the difference between 3

diameters B and B2) of the socket joint during welding, heat from the weld would expand the pipe longitudinally into the shoulder of the socket forcing the pipe-to-flange weld area to move. This could cause the weld to crack. The Socketweld flange was initially developed for use on small size, high-pressure piping in which both a backside hub weld and an internal shoulder weld was made. This provided a static strength equal to the Slip-On flange with a fatigue strength 1.5 times that of the Slip-On flange. Because the two-welds were labor intensive it became the practice to weld only at the hub of the flange. In doing this it relegated the socketweld flange to be more frequently used for small pipe sizes (NPS 2 and below) in non-highpressure, utility type service piping. The Socketweld flange is not approved above Class 1500. Slip-On Flange

fatigue rate is about 66% less than that of a Weld Neck flange. The Slip-On flange is not approved above Class 1500. Lap Joint Flange

Figure 4 The Lap Joint flange requires a companion lap joint, or Type A stub-end (ref. Fig. 5) to complete the joint. The installer is then able to rotate the flange. This allows for quick bolthole alignment of the mating flange during installation without taking the extra precautions required during prefabrication of a welded flange. Their pressure holding ability is about the same as a Slip-On flange. The fatigue life of a Lap Joint/stub-end combination is about 10% that of a Weld Neck flange, with an initial cost that is a little higher than that of a Weld Neck flange.

Figure 3 Unlike the Socketweld flange, the Slip-On flange allows the pipe to be inserted completely through its hub opening. Two welds are made to secure the flange to the pipe. One fillet (pronounced fill-it) weld is made at the hub of the flange and a second weld is made at the inside diameter of the flange near the flange face. The end of the pipe is offset from the face of the flange by a distance equal to the lesser of the pipe wall thickness or 1/4 plus approximately 1/16. This is to allow for enough room to make the internal fillet weld without damaging the flange face. The Slip-On flange is a preferred flange for many applications because of its initial lower cost, the reduced need for cut length accuracy and the reduction in end prep time. However, the final installed cost is probably not much less than that of a Weld Neck flange. The strength of a Slip-On flange under internal pressure is about 40% less than that of a Weld Neck flange. The The real cost benefit in using a Lap Joint flange assembly is realized when installing a stainless steel or other costly alloy piping system. In many cases the designer can elect to use a stub-end specified with the same material as the pipe, but use a less costly, e.g. carbon steel, Lap Joint Flange. This prevents the need of having to weld a more costly compatible alloy flange to the end of the pipe. Just a quick word about stub-ends; they are actually prefabricated or cast pipe flares that are welded directly to the pipe. They are available in three different types: Type A, (which is the lap-joint stub-end), Type B and Type C (ref. Fig. 5). Type A (Fig 5) is forged or cast with an outside radius where the flare begins. This radius conforms to the radius on the inside of the Lap-Joint flange. The mating side of the flare has a serrated surface. Type B (Fig. 5) is forged or cast without the radius where the flare begins. It is used to accommodate the SlipOn flange or Plate flange as a back-up flange.

Type C (Fig 5) is fabricated from pipe using five suggested methods indicated in ASME B31.3. The most prevalent of these is the machine flare. This is done by placing a section of pipe into a flaring machine, flaring the end of the pipe and then cutting it to length. As you can see in the assembly detail of Fig. 5, stub-end types B & C have no radius at the flare while Type A does. This allows it to conform to the Lap-Joint flange. Due to the radius of the type A stub-end, a slip-on flange would have a poor fit, creating non-uniform loading of the flare face as well as an undesirable point load at the radius of the flare. Weld Neck Flange

Figure 7 While the Blind flange is used to cap off the end of a pipeline or a future branch connection it is also used for other purposes. It can be drilled and tapped for a threaded reducing flange or machined out for a Slip-On reducing flange. The reduced opening can be either on-center or eccentric. Flange Pressure Ratings ASME B16.5 flange pressure ratings have been categorized into material groupings. These groupings are formulated based on both the material composition and the process by which the flange is manufactured.

Figure 6 The reinforcement area of the Weld Neck flange distinguishes it from other flanges. This reinforcement area is formed by the added metal thickness, which tapers from the hub of the flange to the weld end. The bore of the flange needs to be specified in order to obtain the same wall thickness at the weld end as the pipe it will be welded to. This will give it the same ID bore as the pipe. The Weld Neck flange is actually the most versatile flange in the ASME stable of flanges. Much of its use is for fitting-to-fitting fabrication in which the flange can be welded directly to a fitting, such as an elbow, without the need for a short piece of pipe, as would be required with a Slip-On flange. It can be used in low-pressure, nonhazardous fluid services as well as high-pressure, highcyclic and hazardous fluid services. While the initial cost of the Weld Neck flange may be higher than that of a Slip-On flange the installed cost reduces that differential. And for conditions of possible high thermal loading, either cryogenic or elevated temperatures, the Weld Neck flange would be essential. Blind Flange

The available pressure Classifications under ASME B16.5 are: 150, 300, 400, 600, 900, 1500 and 2500. The correct terminology for this designation is Class 150, Class 300, etc. The term 150 pound, 300 pound, etc. is a carry over from the old ASA (American Standards Association) Classification. ASA is the precursor to the American National Standards Institute (ANSI). Taking a quick step back, ANSI was founded as a committee whose responsibility was to coordinate the development of standards and to act as a standards traffic cop for the various organizations that develop standards. Its basic function is not to develop standards, but rather to provide accreditation of those standards Originating as the American Engineering Standards Committee (AESC) in 1918, ANSI had, over its first ten years, outgrown its Committee status and in 1928 was reorganized and renamed as the American Standards Association (ASA). In 1966 the ASA was reorganized again under the name of the United States of America Standards Institute (USASI). In 1969 ANSI adopted its present name. While the B16 and B31 Standards have previously carried the ASA and ANSI prefix with its various Standards titles ASME has always been the administrative sponsor in the development of those standards. In the 1970s the prefix designation changed to ANSI/ASME and finally to ASME. Referring to ANSI B16.* or ANSI B31.* is no longer correct. Instead it is correct to refer to a standard as ANSI/ASME B16.* in that it indicates an

ANSI accredited ASME standard. Or you can simply refer to the standard as ASME B16.* or ASME B31.*. Development of ASME B16.5 began in 1920. In 1927 the American Tentative Standard B16e was approved. This eventually became what we know today as ASME B16.5. Until the 1960s the pressure Classifications, as addressed earlier, were referred to as 150 pound, 300 pound, etc. It was at this point the pressure Classification was changed to the Class designation. These designations have no direct correlation with pounds of pressure. Rather, they are a factor in the pressure rating calculation found in B16.5. In Part II of this series, we will discuss how these designations are factored into the design of the flange. Flange Pressure Ratings Flanges, whether manufactured to ASME (American Society of Mechanical Engineers), API (American Petroleum Institute), MSS (Manufacturers Standardization Society), AWWA (American Water Works Association) or any other Standard, are grouped into pressure ratings. In

ASME these pressure ratings are a sub-group of the various material groups designated in B16.5. Figure 8 represents one of the Tables from the Table 2 series in ASME B16.5. This is a series of Tables that lists the Working Pressures of flanges based on material groupings, temperature and Classification. There are 34 Tables segregated into three material Categories of Carbon and low alloy steels, austenitic stainless steels, and nickel alloys. These are further segregated into more defined material sub-groups. Figure 8 shows Table 2-1.1, which indicates, in reverse sequence, sub-category 1 of material group 1 (carbon and low alloy steels). If you had an ASME B16.5, Class 150, ASTM A105 flange this is the table you would use to determine the Working Pressure limit of the flange. To find the Working Pressure of the above mentioned flange enter the column of this table designated as 150 then move down the column to the operating temperature. For intermediate temperatures,

linear interpolation is permitted. In the previous paragraph you will notice that I indicated operating temperature when looking to determine the Working Pressure of a flange. Operating and working are synonymous. The indication of a working pressure and temperature of a fluid service is the same as indicating the operating pressure and temperature. There exists some confusion in this area. That confusion becomes apparent when the engineer is determining design pressure and temperature and applying that to the flange rating. On the surface there appears to be a conflict between rating a flange for design conditions when Table 2 only indicates working pressures. Operating and design pressures and temperatures will be explained in more detail in Article 2. For now I will explain 7

that every service should have an operating pressure/temperature and a design pressure/temperature. A design condition is the maximum coincidental pressure and temperature condition that the system is expected or allowed to see. This then becomes the condition to which you should design for, and to which the leak test is based on, not the operating condition. Tables 2, as it indicates, represents the working or operating pressures of the flange at an indicated temperature for a specific Class. The maximum hydrostatic leak test pressure for a Class 150 flange in Table 2-1.1 is 1.5 times the rated working pressure at 100F, or 285 x 1.5 = 427.5 rounded off to the next higher 25 psi, or 450 psig. We can extrapolate that piece of information to say that since hydrostatic leak test pressure is based on 1.5 x design

pressure the working pressure limit given in the Table 2 matrix ostensibly becomes the design pressure limit. When working with ASME B31.3 Category D fluid services, and initial service leak testing is performed, the working pressure limit then remains the working pressure limit because testing is performed at operating or working pressures. In saying that however, there are caveats that address the fact that not all Category D fluid services should waive the hydrostatic leak test for an initial service leak test. These conditions, such as steam service, will also be discussed in a subsequent article. Category D fluid services are those fluid services that are nonflammable, nontoxic and not damaging to human tissue. Category D fluids additionally do not exceed 150 psig and 366 F. In initial service leak testing the test fluid is the service fluid. Leak testing occurs during or prior to initial operation of the system. As the service fluid is introduced to the piping system and brought to operating pressure, in pressure increments, all joints are observed for possible leaks. If no leaks are detected the pipeline simply remains in service. Other ASME B31.3 fluid services may be expected to operate at one set of conditions, but are designed for another set. For those systems, which might include periodic steamout (cleaning, sterilization, sanitization) or passivation, you therefore want to base your flange rating selection on those more extreme, periodic design conditions. To clarify periodic in this context, the sanitization process can be done as frequently as once per week and last for one to one and half shifts in duration. Flange Facing & Surface Finishes Standard flange facing designations (ref. Fig. 9) are as follows: Flat Face, Raised Face, Ring Joint, Tongue and Groove, Large and Small Male and Female, Small Male and Female (on end of pipe), Large and Small Tongue and Groove. The height of the raised face for Class 150 and 300 flanges is 0.06. The height of the raised face for Class 400 and above is 0.25. Across industry, not discounting the lap-joint flange and stub-end combination, the two most widely used flange facings are the flat face and the raised face. The surface finish of standard raised face and flat face flanges has a serrated concentric or serrated spiral surface finish with an average roughness of 125 in to 250 in. The cutting tool used for the serrations will have a 0.06 in. or larger radius and there should be from 45 to 55 grooves per inch. BOLTS, NUTS & GASKETS

Sealing the flange joint, and as you will see further in this article, the hygienic clamp joint, is paramount in providing integrity to the overall piping system. This is achieved with the use of bolts, nuts and gaskets. Making the right selection for the application can mean the difference between a joint with integrity and one without. ASME B16.5 provides a list of appropriate bolting material for ASME flanges. The bolting material is grouped into three strength categories; high, intermediate and low, which are based on the minimum yield strength of the specified bolt material. The High Strength category includes bolt material with a minimum yield strength of not less than 105 ksi. The Intermediate Strength category includes bolt material with a minimum yield strength of between 30 ksi and 105 ksi. The Low Strength category includes bolt material with a minimum yield strength no greater than 30 ksi. As defined in ASME B16.5, the High Strength bolting materials "may be used with all listed materials and all gaskets". The Intermediate Strength bolting materials "may be used with all listed materials and all gaskets, provided it has been verified that a sealed joint can be maintained under rated working pressure and temperature. The Low Strength bolting materials "may be used with all listed materials but are limited to Class 150 and Class 300 joints", and can only be used with selected gaskets as defined in ASME B16.5. ASME B31.3 further clarifies in para. 309.2.1, "Bolting having not more than 30 ksi specified minimum yield strength shall not be used for flanged joints rated ASME B16.5 Class 400 and higher, nor for flanged joints using metallic gaskets, unless calculations have been made showing adequate strength to maintain joint tightness". B31.3 additionally states in para. 309.2.3, If either flange is to the ASME B16.1 (cast iron), ASME B16.24 (cast copper alloy), MSS SP-42 (valves with flanged and buttweld ends), or MSS SP-51 (cast flanges and fittings) specifications, the bolting material shall be no stronger than low yield strength bolting unless: (a) both flanges have flat faces and a full face gasket is used: or, (b) sequence and torque limits for bolt-up are specified, with consideration of sustained loads, displacement strains, and occasional loads (see paras. 302.3.5 and 302.3.6), and strength of the flanges. In specifying flange bolts, as well as the gasket, it is necessary, not only to consider design pressure and temperature, but fluid service compatibility, the critical nature of the fluid service and environmental conditions all in conjunction with one another. To better understand the relationship of these criteria I will list and provide some clarification for each:

1.

2. 3.

4.

The coincident of design pressure and temperature is what determines the pressure Class of a flange set. That, in turn, along with flange size, will determine the number and size of the flange bolts. The flange Class will also determine the compressibility range of the gasket material. Fluid service compatibility will help determine the gasket material. The critical nature of the fluid will determine the degree of integrity required in the joint. This will help determine bolt strength and material as well as gasket type. Environmental conditions will also help determine bolt material (Corrosive atmosphere, wash-down chemicals, etc.).

not need to specify the nut since it is already defined in A307. However, ASTM A193, alloy and stainless steel bolts, goes only so far when it states that nuts shall conform to ASTM A194, but there are several grades of A194 nuts to select from. This is an example of where the matching nut is not always explicitly called out in the ASTM Standard. Because the ASTM Standards are inconsistent in that regard the spec-writer must make sure it is covered in a specification. You can see from this bit of information that all four components, flanges, bolts, nuts and gaskets have to be selected in conjunction with one another in order for the joint assembly to perform in a way that it is expected to for a given application. Pipe, Tube & Fittings One of the big differences between pharmaceutical and semi-conductor piping and other industrial piping, is the requirements of high purity, or hygienic fluid services. These requirements, as dictated by current Good Manufacturing Practices (cGMP) and defined and quantified by the ISPE and ASME-BPE, are stringent with regard to the manufacture, documentation, fabrication, installation, qualification, validation and quality control of hygienic piping systems and components. The man-hours required in generating, maintaining and controlling the added documentation required for hygienic fabrication and installation is in the range of 30% to 40% of the overall cost of fabrication and installation. Part II in this series will get more into the requirements of hygienic fabrication and where that added cost comes from. For now we will stay with general pipe and fittings. In an attempt at keeping this article concise we will only cover those fittings that are predominantly used throughout industry, both in process and in utility services. Pipe fittings are manufactured by the following processes: cast, forged and wrought. Cast Fittings Cast fittings are provided in cast iron, malleable iron, steel, stainless steel, brass, bronze, and other alloy material as follows: Cast Iron: Cast iron threaded fittings, under ASME B16.4, are available in Class 125 and Class 250 for sizes NPS 1/4 through 12. Cast iron flanged fittings, under ASME B16.1, are available in Class 25, 125 and 250 in sizes NPS 1 through 48. Malleable Iron: Malleable iron fittings, under ASME B16.3, are available in Class 150 and Class 300 in sizes NPS 9

What this ultimately means is that all of the variables that come together in making up a flange joint have to do so in a complementary fashion. Simply selecting a gasket based on material selection and not taking into account the pressure rating requirement could provide a gasket that would get crushed under necessary torque requirements rather than withstand the bolt load and create a seal. Selecting a low strength bolt to be used with a Class 600 flange joint with proper gasketing will require the bolts to be torqued beyond their yield point, or at the very least beyond their elastic range. To explain this briefly; bolts act as springs when they are installed and loaded properly. In order for the flange joint to maintain a gasket seal it requires dynamic loading. Dynamic loading of flange bolts allows expansion and contraction movement in and around the joint while maintaining a seal. This is achieved by applying sufficient stress to the bolt to take it into the materials elastic range. If the bolts are not stressed sufficiently into their elastic range any relaxation in the gasket could reduce the sealing ability of the joint. To the other extreme, if the bolts were stressed beyond their elastic range and into the plastic range of their material of construction the same issue applies, they will lose their dynamic load on the gasket. In this case, if they do not shear they will take a set. Any relaxation in the gasket will then result in the reduction or elimination of the joints sealing ability. With regard to the nut, it should be selected to compliment the bolt. Actually the bolt material specification will steer you, either partially or completely, into the proper nut selection. ASTM A307, a material standard for bolts in the lowstrength category, states that the proper grade for bolts to be used for pipe flange applications is Grade B. A307 goes further to state that when used for pipe flanges Grade B bolts require a Heavy Hex Grade A nut under ASTM A563. In writing a pipe spec that includes the A307 bolt you would

1/8 though 6 for Class 150 and 1/4 through 3 for Class 300. It needs to be noted here that Classifications such as 150 and 300 are not universal throughout the ASME Standards. They are specific to the Standard that they are associated with. You cannot automatically transfer the pressure/temperature limits of a flange joint in ASME B16.5 to that of a fitting in B16.3. Cast Steel: Cast steel, stainless steel and alloy steel flanged fittings, under ASME B16.5, are available in Class 150, 300, 400, 600, 900, 1500 & 2500 in sizes 1/2 though 24. Cast Brass: Cast Brass and bronze threaded fittings, under ASME B16.15, are available in Class 125 and 250, in sizes NPS 1/8 through 4 for Class 125 and 1/4 through 4 for Class 250. Cast Copper: Cast copper solder joints, under ASME B16.18, are available in sizes 1/4 through 6. Forged Fittings Before getting into forged fittings I would like to explain the difference between forged and wrought fittings. There seems to be some vague misconception of what the term forged means and what the term wrought means and how it applies to pipe fittings. The term forging actually comes from the times when metal was worked by hand. A bar of steel would be placed into a forge and heated until it reached its plastic state, at which time the metal would be pulled out of the forge and hammered into some desired shape. Today forging metal basically means working the metal by means of hydraulic hammers to achieve the desired shape. As a small bit of trivia, up until the late 1960s, when mills stopped producing it, wrought iron was the choice of ornamental iron workers. It is still produced in Europe, but most of what we see manufactured as wrought iron in the U.S. is actually various forms of steel made to look like wrought iron. True wrought iron is corrosion resistant, has excellent tensile strength, welds easily and in its plastic range is said to be like working taffy candy. What gives wrought iron these attributes is the iron silicate fibers, or slag added to the molten iron with a small percentage of carbon, whereas cast iron, with a high carbon content, is more brittle and not as easily worked. The smelters, where the iron ore was melted to produce wrought iron, were called bloomeries. In a bloomery the process does not completely melt the iron ore, rather the semi-finished product was a spongy molten mass called a bloom, derived from the red glow of the molten metal, 10

which is where the process gets its name. The slag and impurities were then mechanically removed from the molten mass by twisting and hammering which is where the term wrought originates. Today forged and wrought are almost synonymous. If we look in ASTM A234 - Standard Specification for Piping Fittings of Wrought Carbon Steel and Alloy Steel for Moderate and High Temperature Service we can see in Para 4.1 and in Para 5.1 that wrought fittings made under A234 are actually manufactured or fabricated from material preformed by one of the methods listed previously, which includes forging. In ASTM A961 - Standard Specification for Common Requirements for Steel Flanges, Forged Fittings, Valves and Parts for Piping Applications the definition for the term Forged is, the product of a substantially compressive hot or cold plastic working operation that consolidates the material and produces the required shape. The plastic working must be performed by a forging machine, such as a hammer, press, or ring rolling machine, and must deform the material to produce a wrought structure throughout the material cross section. The difference therefore between forged and wrought fittings is that forged fittings, simply put, are manufactured from bar, which while in its plastic state is formed into a fitting with the use of a hammer, press or rolling machine. Wrought fittings, on the other hand, are manufactured from killed steel, forgings, bars, plates and seamless or fusion welded tubular products that are shaped by hammering, pressing, piercing, extruding, upsetting, rolling, bending, fusion welding, machining, or by a combination of two or more of these operations. In simpler terms wrought signifies worked. There are exceptions in the manufacture of both, but that is the general difference. Something worth noting at this point concerns the ASTM specifications. In quoting from ASTM A961 I was quoting from what ASTM refers to as a General Requirement Specification, which is what A961 is. A General Requirement Specification is a specification that covers requirements that are typical for multiple individual Product Specifications. In this case the individual Product Specifications covered by A961 are A105, A181, A182, A360, A694, A707, A727 and A836. The reason I point this out is that many designers and engineers are not aware that when reviewing an A105 or any of the other ASTM individual Product Specifications you may need to include the associated General Requirement Specification in that review. Reference to a General Requirement Specification can be found in the respective Product Specification. Forged steel and alloy steel socketweld and threaded fittings, under ASME B16.11, are available in sizes NPS 1/8 through 4. Forged socketweld fittings are available in pressure rating Classes 3000, 6000 and 9000. Forged

threaded fittings are available in pressure rating Classes 2000, 3000 and 6000. What I see quite often, and this includes all of the industries I have been associated with, is a misapplication of pressure rating in these fittings. This leads me to believe that the person specifying components does not fully understand the relationship between the pressure Class of these fittings and the pipe they are to be used with. In ASME B16.11 is a table that associates, as a recommendation, fitting pressure Class with pipe wall thickness, as follows:
Table 1 Correlation of Pipe Wall Thickness & Pressure Rating

under B16.28 (short radius 1D elbows), are available in sizes 1/2 through 24. There is no pressure/temperature rating classification for these fittings. In lieu of fitting pressure classifications both B16.9 and B16.28 require that the fitting material be the same as or comparable to the pipe material specification and wall thickness. Under ASME B16.9, given the same material composition, the fittings will have the same allowable pressure/temperature as the pipe. ASME requires that the fittings under B16.28, short radius elbows, be rated at 80% of that calculated for straight seamless pipe of the same material and wall thickness. These fittings can be manufactured from seamless or welded pipe or tubing, plate or forgings. Laterals, because of the elongated opening cut from the run pipe section are rated at 40% of that calculated for straight seamless pipe of the same material and wall thickness. If a full strength lateral is required either the wall thickness of the lateral itself can be increased or a reinforcement pad can be added at the branch to compensate for the loss of material at the branch opening. Wrought copper solder joint fittings, under ASTM B88 and ASME B16.22, are available in sizes 1/4 through 6. These fittings can be used for brazing as well as soldering. The pressure/temperature rating for copper fittings are based on the type of solder or brazing material and the tubing size. It will vary too, depending on whether the fitting is a standard fitting or a DWV (Drain, Waste, Vent) fitting, which has a reduced pressure rating. As an example, using alloy Sn50, 50-50 Tin-Lead Solder, at 100F, fittings 1/2 through 1 have a pressure rating of 200 psig and fittings 1 through 2 have a pressure rating of 175 psig. DWV fittings 1 through 2 have a pressure rating of 95 psig. Using alloy HB, which is a Tin-Antimony-SilverCopper-Nickel (Sn-Sb-Ag-Cu-Ni) solder, having 0.10% maximum Lead (Pb) content, at 100F, fittings 1/2 through 1 have a pressure rating of 1035 psig and fittings 1 through 2 have a pressure rating of 805 psig. DWV fittings 1 through 2 would have a pressure rating of 370 psig. As you can see, within the same type of fitting, there is a significant difference in the pressure ratings of soldered joints depending on the type of filler metal composition. Much of the difference is in the temperature at which the solder or brazing filler metal fully melts. This is referred to as its liquidus state. The temperature at which it starts to melt is referred to as its solidus temperature, the higher the liquidus temperature the higher the pressure rating of the joint. Pipe and Tubing The catch-all terminology for pipe and tubing is tubular products. This includes pipe, tube and their respective 11

Pipe Wall Thk. 80 or XS 160 XXS

Threaded 2000 3000 6000

Socketweld 3000 6000 9000

The ASME recommendation is based on matching the I.D. of the barrel of the fitting with the I.D. of the pipe. The shoulder of the fitting (the area of the fitting that the end of the pipe butts against), either socketweld, as shown in Fig. 10, or threaded, is approximately, allowing for fabrication tolerances, the same width as the specified mating pipe wall thickness.

Figure 10 Socket Weld Fitting Joint from ASME B16.11 As an example, referring to Table 1, if you had a specified pipe wall thickness of Sch. 160 the matching threaded forged fitting would be a Class 3000, for socketweld it would be a Class 6000. The fitting pressure Class is selected based on the pipe wall thickness. Referring to Fig. 10, you can readily see that by not matching the fitting Class to the pipe wall thickness it will create either a recessed area or a protruding area the length of the barrel of the fitting, depending on which side you error on. For forged reinforced branch fittings refer to MSS Standard SP-97 Integrally Reinforced Forged Branch Outlet Fittings Socket Welding, Threaded and Buttwelding Ends. Wrought Fittings Wrought Steel Butt Weld Fittings under ASME B16.9 (standard radius 1.5D elbows and other fittings) are available in sizes 1/2 through 48. Wrought Steel Butt Weld Fittings

fittings. Piping itself refers to a system of pipe, fittings, flanges, valves, bolts, gaskets and other in-line components that make up an entire system used to convey a fluid. The simple distinction between pipe and tubing is that tubing is thin-walled pipe with a different size for size diameter. Tubular products can basically be grouped into three broad classifications: pipe, pressure tube and mechanical tube. Based on user requirements the above classifications come in various types such as Standard Pipe, Pressure Pipe, Line Pipe, Water Well Pipe, Oil Country Tubular Goods, Conduit, Piles, Nipple Pipe and Sprinkler Pipe. The two types that we are mainly interested in are Standard and Pressure Pipe. Distinguishable only from the standpoint of use, Standard Pipe is intended for low pressure, non-volatile use, whereas Pressure Pipe is intended for use in higher integrity services. These are services in which the pipe is required to convey high pressure volatile or non-volatile liquids and gases at sub-zero or elevated temperatures. The following represents a combined description of Standard and Pressure Pipe. Pipe: Pipe is manufactured to a NPS in which the OD of a given nominal size remains constant while any change in wall thickness is reflected in the pipe ID. Pipe wall thicknesses are specified by Schedule (Sch.) numbers 5, 10, 20, 30, 40, 60, 80, 100, 120, 140 and 160. Add the suffix s when specifying stainless steel or other alloys. Wall thickness is also specified by the symbols Std. (Standard), XS (Extra Strong) and XX (Double Extra Strong). Pipe NPS 12 and smaller has an OD that is nominally larger than that specified. Pipe with a NPS 14 and larger has an OD equal to the size specified. Tubing: Steel and alloy tubing is manufactured to an OD equal to that specified. Meaning that 1/4 tubing will have a 1/4 OD, 2 tubing will have a 2 OD. Copper tubing, accept for ACR (Air-Conditioning & Refrigeration) tubing, which has an OD equal to that specified, has an OD that is always 1/8 larger than the diameter specified. As an example, 1/2 tubing will have a 5/8 OD, 1 tubing will have a 1 1/8 OD. Wall thickness for tubing is specified in the actual decimal equivalent of its thickness. Pipe is manufactured in three basic forms: cast, welded and seamless. Tubing is manufactured in two basic forms: welded and seamless. Cast Pipe: Cast pipe is available in four basic types: white iron, malleable iron, gray iron and ductile iron. White iron has a high carbon content in the carbide form. The lack of graphite gives it its light colored appearance. Carbides 12

give it a high compressive strength and a hardness that provides added resistance to wear, but leave it very brittle. Malleable iron is white cast iron that has been heat treated for added ductility. By reheating white cast iron in the presence of oxygen containing materials such as iron oxide, and allowing it to cool very slowly, the free carbon forms small graphite particles. This gives malleable iron excellent machinability and ductile properties along with good shock resistant properties. Gray iron is the oldest form of cast iron pipe and is synonymous with the name cast iron. It contains carbon in the form of flake graphite, which gives it its gray identifying color. Gray cast iron has virtually no elastic or plastic properties, but has excellent machining and self-lubricating properties due to the graphite content Ductile iron is arguably the most versatile of the cast irons. It has excellent ductile and machinable properties while also having high strength characteristics. Welded Steel Pipe and Tubing: Referring to pipe in the following also includes tubing. Welded steel pipe is manufactured by Furnace Welding or by Fusion Welding. Furnace Welding is achieved by heating strip steel, also referred as skelp, to welding temperature then forming it into pipe. The continuous weld, or buttweld, is forged at the time the strip is formed into pipe. This is a process generally used to manufacture low cost pipe 3 and below. Fusion Welded pipe is formed from skelp that is cold rolled into pipe and the edges welded together by resistance welding, induction welding or arc welding. Electric resistance welding (ERW) can be accomplished by flash welding, high-frequency or low-frequency resistance welding. A scarfing tool is used to remove upset material along the seam of flash-welded pipe. Flash welding produces a high strength steel pipe in NPS 4 through 36. Low-frequency resistance welding can be used to manufacture pipe through NPS 22. Highfrequency resistance welding can be used to manufacture pipe through NPS 42. High-frequency induction welding can be used for high rate production of small NPS pipe. This is a cleaner form of welding in which scarfing, or the cleaning of upset material along the seam, is normally not required. Arc welding the longitudinal seam of production pipe is accomplished with submerged arc welding (SAW), inert gas tungsten arc welding (GTAW) also called tungsten inert gas welding (TIG), or gas shielded consumable metal arc welding (MIG).

As you will see in Part II, the type of weld seam used in the manufacture of pipe is a factor when calculating the Pressure Design Thickness (t) of the pipe wall. It reduces the overall integrity of the pipe wall by a percentage given in ASME B31.3 based on the type of longitudinal seam weld. Seamless Steel Pipe and Tubing: Referring to pipe in the following also includes tubing. Seamless steel pipe, using various extrusion and mandrel mill methods, is manufactured by first creating a tube hollow from a steel billet, which is a solid steel round. The billet is heated to its hot metal forming temperature then pierced by a rotary piercer or by a press piercer creating the tube hollow, which will have a larger diameter and thicker wall than its final pipe form. The tube hollow is then hotworked by the Mandrel Mill Process, Mannesmann PlugMill Process, or Ugine Sejournet Extrusion Process. Upon completion of these processes the pipe is referred to as hot-finished. If further work is required to achieve more accuracy in the diameter, wall thickness or improve its finish the pipe can be cold-finished, or cold-worked. When the pipe is cold-finished it will require heat treating to remove stress in the pipe wall created when worked in its cold state. There are also two forging processes used in the manufacture of large diameter (10 to 30 inch) pipe with heavy wall thickness (1.5 to 4 inch). The two forging methods are called Forged and Bored, and Hollow Forged. Other Material and Systems We have touched on just some of the key points of steel pipe and fittings. What I have not touched on are plastic lined pipe systems and non-metallic piping including proprietary piping systems. The area of non-metallic piping is certainly worth including in the context of piping. However, we will keep these articles focused on metallic piping material. Non-metallic piping merits a discussion on its own, and should not be relegated to a paragraph or two here. However, since plastic lined pipe is steel pipe with a liner and is so widely used in the various industries I will touch on some of its key points. Lined Pipe Systems: Lined flex hoses were first developed in 1936 by Resistoflex followed by lined pipe, which did not come to the industry until 1956 by way of the same company. When first introduced, plastic line pipe filled a large fluid handling gap in industry, but brought with it some technical issues. As other manufacturers such as Dow and Peabody Dore began producing lined pipe and fittings industry standards for lined pipe did not exist. Consequently, there 13

were no standard fitting dimensions and the availability of size and type of fittings would vary from one company to another, and still, to a much lesser degree, does to this day. Due to the autonomous nature of lined pipe manufacturing during its initial stages the pipe designer would have to know early in the design process which manufacturer they were going to use. Particularly in fitting make-up situations, you needed to know in advance what those make-up dimensions were going to be, and thus the fitting manufacturer. While not having industry standard dimensions was a design problem other operational type problems existed as well. Some of the fluid services these line pipe systems were specified for, and still are, would normally be expected to operate under a positive pressure, but at times would phase into a negative pressure. The liners in these early systems were not necessarily vacuum rated and at times would collapse under the negative internal pressure, plugging the pipeline. There was an added problem when gaskets were thrown into the mix. Gaskets were not normally required unless frequent dismantling was planned, and many firms, both engineers and manufacturers, felt more secure in specifying gaskets at every joint. When required, the gasket of choice, in many cases, was an envelope type gasket made of PTFE (polytetrafluoroethylene) with an inner core of various filler material,Viton (a DuPont trade name) or EPDM. These gaskets had a tendency to creep under required bolt torque pressure at ambient conditions. From the time a system was installed to the time it was ready to hydrotest the gaskets would, on many occasion, creep, or relax to the point of reducing the compressive bolt load of the joint enough to where it would not stand up to the hydrotest pressure. Quite often leaks would become apparent during the fill cycle prior to testing. There also exists the problem of permeation with regard to PTFE liner material and of Internal and External Triboelectric Charge Generation and Accumulation (static electricity). But, due to the diligent efforts of the line pipe and gasket industries these types of problems have either been eliminated or controlled, and some are still being pursued. Fitting dimensions have been standardized through ASTM F1545 in referencing ASME B16.1 (cast iron fittings), B16.5 (steel fittings) and B16.42 (ductile iron fittings). You will need to read Note 3 under Sub-Para. 4.2.4, which states, Center-to-face dimensions include the plastic lining. Meaning, the dimensions given in the referenced ASME standards are to the bare metal face of the fittings. However, when lined fittings are manufactured the metal casting is modified to accommodate the liner thickness being included in that same specified center-to-face dimension.

With regard to vacuum rating, liner specifications are greatly improved, but you will need to check the vacuum ratings of available pipe and fittings with each tentative manufacturer. This provision will vary from manufacturer to manufacturer depending on size, fitting, liner type, pressure and temperature. Gasket materials such as Garlocks Gylon gasket, which is a PTFE/Silicate composite, and W. L. Gores Universal Pipe Gasket, which is a 100% expanded PTFE, have been developed to reduce the creep rate in a gasket material that is compatible with virtually the same fluid services that lined pipe systems are usually selected for. Permeation issues with PTFE liners (it also exists, to a lesser extent, with other liner material) have been accommodated more than resolved with the use of vents in the steel pipe casing, the application of vent components at the flange joint, and increased liner thickness. Internal and external charge accumulation, known as static electricity, or triboelectric charge accumulation, is the result of an electrical charge generation unable to dissipate. If the electrical charge generation is allowed to continually dissipate to ground then there is no charge build-up and no problem. This is what occurs with steel pipe in contact with a flowing fluid. Charge generation has a path to ground and does not have an opportunity to build up. With regard to thermoplastic lined pipe there are two issues to be considered: external charge accumulation and internal charge accumulation. This is an issue that requires experience and expertise in order to analyze a particular situation. What we will do in Part II of this series is provide you with basic information that will at least allow you to be familiar with the subject, and help you to understand the issues. Standard sizes of plastic lined pipe and fittings range from NPS 1 through 12. Edlon, a lined pipe manufacturer, also manufactures larger diameter pipe and fittings from NPS 14 through 24, and when requested can manufacture spools to 144 diameter. Hygienic Piping Hygienic is a term defined in ASME-BPE as: of or pertaining to equipment and piping systems that by design, materials of construction, and operation provide for the maintenance of cleanliness so that products produced by these systems will not adversely affect animal or human health. While system components such as tube, fittings, valves and the design itself, with regard to hygienic conditions, can translate to the Semi-Conductor industry the term hygienic 14

does not. It pertains strictly to the health aspects of a clean and cleanable system for the pharmaceutical industry. The semi-conductor industry requires a high, or in some cases higher, degree of cleanliness and cleanability than hygienic systems in the pharmaceutical industry, for altogether different reasons. A term that can more appropriately be interchanged between these two industries is high-purity. This implies a high degree of cleanliness and cleanability without being implicitly connected with one industry or the other. For what is referred to as product contact material, the surface roughness, dead-leg minimums and an easily cleanable system are all imperative. Because of this the pharmaceutical industry had to make a departure from the 3A standards it plagiarized early on in order to develop a set of guidelines and standards that better suit its industry. Enter ASME-BPE. ASME-BPE has taken on the task of providing a forum for engineers, pharmaceutical manufacturers, component and equipment manufacturers, and inspectors in an effort to develop consensus standards for the industry where none existed before. I wont go further with this except to say that, to the handful of engineers undaunted by the task ahead of them, in approaching ASME about the need to create another standards committee, and the perseverance to see it through; my hat goes off to you. Hygienic piping was, up until just recently, referred to as sanitary piping. Because this term has been so closely associated with the plumbing industry and sanitary drain piping it is felt by the pharmaceutical industry that the change in terminology to hygienic is more appropriate. In both the pharmaceutical and semiconductor industries the need for crevice free, drainable systems is a necessity. This translates into weld joint quality, mechanical joint design requirements, interior pipe surface roughness limits, system drainability and dead-leg limitations. Slope, welding, dead-legs and surface roughness will be discussed in Part II. This article will concentrate on the basic aspects of the fittings. Fittings There are two basic types of fitting joints in hygienic piping: welded and clamp. Welded fittings, unlike standard buttweld pipe fittings, have an added tangent length to accommodate the orbital welding machine. The orbital welding machine allows the welding operator to make consistent high quality autogenous welds. Autogenous welds are welds made without filler metal. Fusion is made between the parent metals of the two components being welded by means of tungsten inert gas welding; more on welding in Part II.

hold-up. This can result in the loss of residual product, cause potential cross-contamination of product, and promote microbial growth. The reason I mention this here, and I wont go into it any further until Part II, is because there are manufacturers that are attempting to overcome these issues by improving on the concept of the clamp joint. Two manufactures, Swagelok and The Neumo Ehrenberg Group, represented in the US by VNE, have, what I would consider, well developed re-designs of the standard hygienic clamp assembly.

Figure 11 - Fittings Ready To Be Orbital Welded Compliments of ARC Machines, Inc. Figure 11 is an example of an orbital, or automatic, welding machine mounted on its work-piece. In this example it happens to be a 90 elbow being welded to a cross. You can see by this example why the additional straight tangent section of automatic weld fittings is needed. That extra length provides a mounting surface for attaching the automatic welding machine. The clamp connection is a mechanical connection whose design originated in the food and dairy industry, but whose standardization has been under development by ASME-BPE. Due to a lack of definitive standardization most companies that use this type connection require in their specifications that both the ferrule, the component that the clamp fits on, and the clamp itself come from the same manufacturer. This is to ensure a competent fit. There are no specific dimensions and tolerances for the clamp assembly, except for that which is being developed by ASME-BPE. Currently it is possible to take a set of ferrules from one manufacturer, mate them together with a gasket, attach a clamp from a different manufacturer and tighten up on the clamp nut. In some cases you can literally rotate the clamp by hand about the ferrules. Meaning, there is no force being applied on the joint seal. For those of you unfamiliar with the clamp joint, it is the clamp that applies the force that holds the ferrules together. The fact that this can occur begs the need for standardization to a greater degree than what currently exists. Well get into this in greater detail in Part II, but another issue that currently exists with the clamp joint is gasket intrusion into the pipe ID due to inadequate compression control of the gasket. Gasket intrusion is a problem in pharmaceutical service for two reasons: 1. Depending on the hygienic fluid service and the gasket material the gasket protruding into the fluid stream can break down and slough off into the fluid flow, contaminating the hygienic fluid. 2. The intrusion of the gasket into pipe ID on a horizontal line can also cause fluid

Figure 12 Swagelok TS Series Profile Compliments Swagelok Company Swagelok has developed what they call their TS series fittings. These ferrules (Fig. 12) have a design that provides compression control of the gasket while also controlling the creep tendency inherent in, arguably, the most prevalent gasket material used in high purity piping, Teflon.

Figure 13 Maxpure Connect S Compliments Neumo Ehrenberg Group The Neumo Ehrenberg Group manufactures a clamp joint (also provided as a bolted connection) that does not require a gasket (Fig. 13). This type of joint, called the Connect-S under their newly formed MaxPure label of fittings, is currently in use in Europe. While this connection alleviates the issues that are present with a gasketed joint 15

added care would need to be applied in its handling. Any scratch or ding to the faced part of the sealing surface could compromise its sealing integrity. Nevertheless this is a connection design worth consideration. In this first article we have covered a few of the basics, which will provide us with a little more insight when we discuss the more in-depth topics of piping Codes, piping design, and fabrication of pipe in Part II. Future Articles What we have discussed so far is just some of the basics of general piping. While there is a great deal left unsaid we will provide further clarification as we move through the next two articles. The next article, titled Piping Design Part II Code, Design and Fabrication, will cover the more specific aspects of Code governance, engineering in pipe design and fabrication as it relates to welding, assembly and installation. The third article in this series, titled Piping Design Part III Installation, Cleaning, Testing and Verification, will wrap up the series by discussing the four title points. Acknowledgement: I wish to thank Earl Lamson, senior Project Manager with Eli Lilly and Company, for taking time out of a busy schedule to read through the draft of this article. He obliged me by reviewing this article with the same skill, intelligence and insight he brings to everything he does. His comments kept me concise and on target.

About the author: W. M. (Bill) Huitt has been involved in industrial piping design, engineering and construction since 1965. Positions have included design engineer, piping design instructor, project engineer, project supervisor, piping department supervisor, engineering manager and president of W. M. Huitt Co. a piping consulting firm founded in 1987. His experience covers both the engineering and construction fields and crosses industrial lines to include petroleum refining, chemical, petrochemical, pharmaceutical, pulp & paper, nuclear power, biofuel, and coal gasification. He has written numerous specifications, guidelines, papers, and magazine articles on the topic of pipe design and engineering. Bill is a member of ISPE (International Society of Pharmaceutical Engineers), CSI (Construction Specifications Institute) and ASME (American Society of Mechanical Engineers). He is a member of three ASMEBPE subcommittees, several Task Groups, an API Task Group, and sets on two corporate specification review boards. He can be reached at: W. M. Huitt Co. P O Box 31154 St. Louis, MO 63131-0154 (314)966-8919 wmhuitt@aol.com www.wmhuitt.com

16

As published in the June and July 2007 issues of Chemical Engineering Magazine

Piping Design Part 2: Code, Design and Fabrication

There is not a reason sufficiently good enough not to comply with appropriate industry Standards and Codes.
W. M. (Bill) Huitt W. M. Huitt Co. A request was put to me a few years back asking if I would respond in writing to the question, Why do we, as a company, need to comply with a piping Code? The question was in regard to the building of industrial facilities, and was in preparation for a meeting that was about to take place for which the main topic was going to be the issue of Code compliance. If you considered the question while reading it you may have noticed that there is, although unintentional, a trick to it. Code, by definition is law with statutory force. Therefore the reason for complying with a Code is because you literally have to, or be penalized for non-compliance. The question actually intended was, why comply with or adopt a piping consensus standard? When a question like the one above is phrased as it is it supports my contention that many people, referring to engineers and designers in our case, do not fully understand the difference between a Code and a Standard. And it doesnt help matters when some Standards are published as a Code, and some Codes are published as a Standard. This is certainly nothing to get excited about, but it is something worth pointing out. My take on the reason for the misunderstanding of these two closely related terms, Standard and Code, is that they get bounced around so often in the same context that designers and engineers simply begin interchanging the two terms without much consideration for their different meanings. Im going to explain the difference between a Standard and a Code, but before I do, heres the written response I gave to the above question: Consensus Standards such as those published by ASME, ANSI, API, NFPA, ASTM, International Plumbing Code and others are not mandatory in and of themselves. However, federal, state, city and other local Codes are mandatory. In these municipal Codes you will find regulations that establish various requirements taken in whole, or in part from the Standards published by the above listed organizations, and others, as legally binding requirements. These Standards, as adopted, then become Code, which is enforceable by law. When not addressed on a municipal level, but included in corporate specifications, the Standard becomes a legal Code on a contractual basis. To comply with these Codes, irrespective of government regulations or corporate requirements, doesn't cost the builder any more than if they didn't comply. It does, however, cost more to fabricate and install piping systems that have a high degree of integrity as opposed to a system that doesn't. By hiring non-certified welders and plumbers, bypassing inspections, examinations and testing, using material that may potentially not withstand service pressures and temperatures, and supporting this type of system with potentially inadequate supports is less costly but there's too much at risk. I don't think anyone in good conscience would

intentionally attempt to do something like that in order to save money, but then again the world is full of unscrupulous individuals and corporations. If anyone intending on fabricating and installing a piping system plans to: 1. Use listed material, 2. Specify material that meets the requirements for fluid service, pressure and temperature, 3. Inspect the material for MOC, size and rating, 4. Use certified welders and plumbers, 5. Inspect welds and brazing, 6. Adequately support the pipe, 7. Test the pipe for tightness; Then they are essentially complying with Code. The Code simply explains how to do this in a formal, well thought-out manner. There is not a reason sufficiently good enough not to comply with appropriate industry Standards and Codes. If there was a fee involved for compliance this might be a stimulus for debate. But there is no fee, and there is usually just too much at stake. Even with utility systems in an admin building or an institutional facility, the potential damage from a ruptured pipeline, or a slow leak at an untested joint could easily overshadow any savings gained in noncompliance. That's without considering the safety risk to personnel. The first thing that someone should do, if they are considering to do otherwise, is check local and state Code. They may find regulations that require adherence to ASME, the International Plumbing Code or some of the other consensus Standards. If not already included, this should be a requirement within any companys specifications. Just a bit of trivia: ASME published the first edition of the Boiler and Pressure Vessel Code in 1914-15. Prior to creation of the Code, and what played a large part in instigating its creation, was that between 1870 and 1910 approximately 14,000 boilers had exploded. Some were devastating to both people and property. Those numbers fell off drastically as the Code was adopted. Uniformity and regulation does have its place. PIPING CODE In a piping facility, defined here as an industrial facility requiring a significant amount (apply your own order of magnitude here) of pipe, the three key factors in its development are the governing Code, the design (includes specifications and engineering), and pipe fabrication (includes installation). These are the three topics we will discuss on a broad, but limited basis in this article. 2

Like the seatbelt law Code compliance is not just the law, it makes good sense. A professional Consensus Standard is, very simply put, a Code waiting to be adopted. Take the ASME Boiler and Pressure Vessel Code (BPVC), since its first publication in 1915 it has been adopted by 49 states, all the provinces of Canada, and accepted by regulatory authorities in over 80 countries. On May 18, 2005 it was finally adopted by the 50 th state, South Carolina. And this doesnt mean the BPVC is adopted in its entirety. A state, or corporation for that matter, can adopt a single section or multiple Sections of the BPVC, or they can adopt it in its entirety. Until South Carolina adopted the BPVC it was actually no more than a Standard in that state and only required compliance when stipulated in a specification. However, in all honesty you would not get a US boiler or pressure vessel manufacturer to by-pass Code compliance. That is, unless you wanted to pay their potential attorneys fees. With regard to Code compliance, the question I get quite often is, How do I determine which piping Code, or Standard, I should comply with for my particular project? Determining proper Code application is relatively straightforward while at the same time providing a certain degree of latitude to the Owner in making the final determination. In some cases that determination is made for the Engineer or Contractor at the state level, the local level or by an Owner company itself. Providing guidelines for Code adoption on a project basis is direction that should be included in any companys set of specifications, but quite often is not. This can cause a number of disconnects through design and construction. In order to answer the question about Code assignment some history has to be told. In keeping this brief I will just touch on the high points. In 1942, ASA B31.1 American Standard Code for Pressure Piping was published by the American Standards Association. This would later change to B31.1 - Power Piping. In the early 1950s the decision was made to create additional B31 Codes in order to better define the requirements for more specific needs. The first of those Standards was ASA B31.8 Gas Transmission and Distribution Piping Systems, which was published in 1955. In 1959 the first ASA B31.3 Petroleum Refinery Piping Standard was published. After some reorganization and organizational name changes the ASA became ANSI, American National Standards Institute. Subsequent Code revisions were designated as ANSI Codes. In 1978, ASME was granted accreditation by ANSI to organize the B31 Committee as the ASME Code for Pressure Piping. This changed the Code designation to ANSI/ASME B31.

Since 1955 the B31 Committee has continued to categorize, create and better define Code requirements for specific segments of the industry. Through the years since then they have created, not necessarily in this order, B31.4 Liquid Transportation Piping, B31.5 Refrigeration Piping, B31.9 Building Services Piping, and B31.11 Slurry Transportation Piping. Each of these Standards is considered a stand-alone Section of the ASME Code for Pressure Piping, B31. What the B31 committee has accomplished, and is continuing to improve upon, are Standards that are better focused on specific segments of industry. This alleviates the need for a designer or constructor building an institutional type facility from having to familiarize themselves with the more voluminous B31.3 or even a B31.1. They can work within the much less stringent and extensive requirements of B31.9, a Standard created for and much more suitable for that type of design and construction. As mentioned above, ASME B31.1 Power Piping, was first published in 1942. Its general scope reads: Rules for this Code Section have been developed considering the needs for applications which include piping typically found in electric power generating stations, in industrial and institutional plants, geothermal heating systems, and central and district heating and cooling systems. The general scope of ASME B31.3 Process Piping, reads: Rules for the Process Piping Code have been developed considering piping typically found in petroleum refineries, chemical, pharmaceutical, textile, paper, semiconductor and cryogenic plants; and related processing plants and terminals. ASME B31.5 Refrigeration Piping, applies to refrigerant and secondary coolant piping systems. Closely related to B31.1, but not having the size, pressure or temperature range, B31.9 was first published in 1982. It was created to fill the need for piping in limited service requirements. Its scope is narrowly focused on only those service conditions that may be required to service the utility needs of operating a commercial, institutional or residential building. From its shear scope of responsibility, B31.3 encompasses virtually all piping, including those also covered by B31.1 (except for boiler external piping), B31.5 and B31.9. The difference, and distinction, as to which Code should apply to a particular project, lies with the definition and scope of the project itself. If a project includes only the installation of perhaps a refrigeration system, B31.5 would apply. If a project's scope of work consists of an office, laboratory, research facility, institutional facility or any combination thereof, B31.1 or B31.9 and possibly B31.5 would apply. A laboratory or 3

research facility could possibly require fluid services beyond the fluid service limits of B31.9. In that case, B31.3 would be adopted for those services. In the case of a process manufacturing facility, B31.3 would be the governing Code. Since B31.3 covers all piping, B31.5 or B31.9 would not need to be included, not even necessarily with associated lab, office and research facilities. The only time B31.5 or B31.9 would become governing Codes, in association with a manufacturing facility, is if a refrigeration unit, or an office, lab and/or research facility were under a separate design/construct contract from the process manufacturing facility. Or they were a substantial part of the overall project. As an example, project XYZ consists of a process manufacturing facility, related office building and lab facilities. If the utility service piping for the office and lab facilities is a small percentage of the overall project, and/or the design and construction contracts for those facilities are a part of the overall process manufacturing facility, all piping, with Code exclusions, could be governed by B31.3. If, however, the office and lab facilities were a substantial part of the overall project, or they were to go to a separate constructor it may be more beneficial to determine battery limits for those facilities and designate anything inside those battery limits as B31.1 or B31.9 and/or B31.5. In such a case, separate pipe specifications may have to be issued for those portions of the project designated as being governed by B31.9. This is due to the range of fluid services and the corresponding pressure and temperature limits of B31.9 compared to those of B31.3. These differences in Code assignment and battery limits may be a driver for the projects contracting strategy. Many piping service requirements such as steam, air, chilled water, etc. can come under the auspices of multiple Codes. These fluid services, which fall within the definition of B31.3 Category D fluid services, can just as easily fall within the requirements of B31.1 or B31.9 as well. In an effort at maintaining a high degree of continuity in the process of making the determination of which Code to apply to a project, company guidelines should be well defined. The final determination as to what constitutes a governing Code, within the purview of the above mentioned Codes, is left to the Owner and/or to the local governing jurisdiction. Engineering specs should clarify and reflect the intent of the Owner and the respective Codes in an attempt to provide consistency and direction across all projects within a company. PIPING DESIGN Piping design is the job of configuring the physical aspects of pipe and components in an effort to conform with P&IDs, fluid service requirements, associated material

specifications, equipment data sheets, and current Good Manufacturing Practice while meeting Owner expectations. All of this has to be done within a pre-determined threedimensional assigned space while coordinating that activity with that of the architecture, structural steel, HVAC, electrical, video, data & security conduit and trays, and operational requirements. Pulling together and coordinating the above mentioned discipline activities to achieve such a compilation of design requires a systematic methodology, planning, technical ability, coordination, foresight, and above all experience. A note of omission here: CAD (Computer Aided Design) is such an integral part of piping design that its difficult, if not impossible, to discuss design without including CAD in the discussion. It plays such a large part that, rather than enter into it here, I will dedicate an entire article to it at a later date. That article will discuss the integration of CAD into the industry including its merits, and how, in many respects, its method of implementation and integration has inversely diminished the quality of design with respect to industrial piping. The article will also discuss industrys reaction to this unexpected result, and the issues we are still dealing with today in the use of CAD. PIPING SPECIFICATIONS One of the first activities the piping engineer will be involved with is development of piping specifications, design guidelines and construction guidelines. Piping specifications, as an overview, should provide essential material detail for design, procurement and fabrication. Guidelines, both design and construction, should provide sufficient definition in a well organized manner to allow the designer and constructor the insight and direction they need in order to provide a facility that will meet the expectation of the Owner with minimal in-process direction from the Owner or Construction Manager. Piping Specifications A Piping Specification is the document that will describe the physical characteristics and specific material attributes of pipe, fittings and manual valves necessary to the needs of both design and procurement. These documents also become contractual to the project and those contractors that work under them. Design will require a sufficient degree of information in a specification that will allow for determining the service limitations of the specification and what fluid services the specifications material is compatible with. That is, a project may have, among other fluid services, sulfuric acid and chilled water. The economic and technical feasibility of the material selection for chilled water service would not be technically feasible for sulfuric acid. Inversely, the economic

and technical material selection for sulfuric acid service would not be economically feasible for chilled water service. Procurement too, will need detailed specifications to limit the assumptions they will have to make or the questions they will have to ask in preparing purchase orders. The piping specification should make clear exactly what the material of construction is for each component, and what standard that component is manufactured to. Also included in the component description should be pressure rating, end connection type and surface finish where required. There are a few rather consistent mistakes that companies make in developing or maintaining specs: 1. within the spec itself they are either not definitive enough or they are too definitive; 2. they are not updated in a timely manner; and/or 3. The specs are too broad in their content. In defining the above issues well begin with: Point #1: When defining pipe and components in a specification you should provide enough information to identify each component without hamstringing yourself or procurement in the process. What I mean by that is, do not get so specific or proprietary with the specification that only one manufacturer is qualified to provide the component, unless you intend to do just that. With standard pipe and fittings its difficult to provide too much information. However, with valves and other inline equipment it can happen quite easily. A common practice of spec writers is to write a specification for a generic type valve, one that can be bid on by multiple potential suppliers, by using the description of one particular valve as a template. What happens is that proprietary manufacturer trade names, such as some of the trim materials, are carried over to the generic valve spec. When the procurement person for the mechanical contractor, or whoever is buying the valves for the project, gets ready to buy this valve the only manufacturer that can supply it with the specified proprietary trim is the one from which the spec was copied. You would think that, in doing this, it would eliminate multiple bids for the valve based on the unintentional proprietary requirements in the spec. In actuality it creates confusion and propagates questions. The valve bidders, other than the one the spec was based on, will bid the valve with an exception to the proprietary material, or they will contact the purchasing agent for clarification. Since the purchasing agent wont have the answer, the question, or actually the clarification, then goes back to the engineer and/or the Owner. The time necessary in responding to these types of issues is better spent on more pressing matters. When developing a spec be specific, but try not to include proprietary data unless you intend to. As an example when specifying Viton you are specifying a generic DuPont 4

product. Generic in that there are several different types of Viton such as Viton A, Viton B, Viton GF, Viton GFLT, etc. Each of these has specific formulations, which gives them different fluid service compatibility and pressure/ temperature ranges. Viton is a type of fluorocarbon. Fluorocarbons are designated FKM under ASTM D-1418. So when specifying Viton you are identifying a specific product from a specific manufactureralmost. If, in developing a specification, you wish to establish minimum requirements for a component or a material it is certainly acceptable to identify a specific proprietary item as a benchmark. In doing this, and well stay with the fluorocarbon gasket or seal material example, you could identify Viton GF or equal, which would indicate that a comparable material from one of the other fluorocarbon manufacturers would be acceptable so long as the fluid service compatibility and pressure/temperature ranges were equal to or greater than the Viton GF material. In saying almost above what I meant by that is, if you write the spec as Viton you would most likely get the original formulation, which is Viton A. The fluid service may be more suited for an FKM with polytetrafluoroethylene in it. That would be a Viton GF. Or an FKM suitable for colder temperatures may be a better choice. That would be a Viton GFLT. Be specific for those that have to use the specs to design from and those that have to purchase the material. Point #2: All too often after a specification is developed it will reside in the companys database without being periodically reviewed and updated. Industry standards change, part numbers change, manufacturers are bought and sold; manufacturers improve their products, etc. All of these things constitute the need and necessity to review and revise specifications on a timely basis. A company that houses their own set of specifications should review those specifications at least every two years. This timing works out for a couple reasons: 1. industry standards, on average, publish every two years, and 2. capital projects, from design through close-out, will arguably have an average duration of two years. Lessons-learned from projects can then be considered for adoption into company specs, prompting a new revision. Point #3: Specs being too broad in their content refers to an attempt at making the specs all-inclusive. A piping specification should contain only those components and information that would typically be used from job to job. That would include the following (as an example): 1. 2. 3. 4. 5. Pressure/Temperature limit of the spec Limiting factor for Pressure/Temperature Pipe material Fitting type, rating and material Flange type, rating and material 5

6. 7. 8. 9. 10.

Gasket type, rating and material Bolt & nut type and material Manual valves grouped by type Notes Branch chart matrix with corrosion allowance

The ten line items above provide the primary component information and notations required for a typical piping system. Some specifications are written to include such components as steam traps, sight glasses, 3-way or 4way valves, strainers, and other miscellaneous type items. Those miscellaneous items are better referred to as specialty items (or some other similar descriptive name) and are sized and specified for each particular application. This does not make them a good candidate for inclusion into a basic pipe specification. To explain the above we can use, as an example, a carbon steel piping system that is specified to be used in a 150 psig steam service. The pipe, flanges, fittings, bolts, gaskets and valves can all be used at any point in the system as specified. The specification for a steam trap, however, will vary depending on its intended application. And depending on its application the load requirements for each trap may vary. As an example, a steam trap application at a drip leg will have a light steady load, whereas a steam trap application at a shell & tube heat exchanger may have a heavier modulating load. And that doesnt take into account the need for the different types of traps, e.g. F&T, inverted bucket, thermodynamic, etc. You could, depending on the size of the project, have multiple variations of the four basic types of steam traps with anywhere from 30 to 300 or more traps in multiple sizes and various load requirements. I think you can see why this type of requirement needs to be its own specification and not a part of the piping specification. A piping specification should be concise, definitive and repeatable. Adding specialty type items to the specification makes it convoluted and difficult to control and interpret. Users of these specifications are designers, bidders, procurement personnel, fabricators, receipt verification clerks, validation and maintenance personnel. With that in mind you can better understand, or at least value the fact, that these documents have to be interpreted and used by a wide range of personnel. Those personnel are looking for particular information, written in a concise manner that will allow them to design and order or verify components within that specification. In attempting to include the specialty type items it will, at the very least, complicate and exacerbate the process. DESIGN AND CONSTRUCTION GUIDELINES

Design and construction guidelines, working in conjunction with the piping specifications, should convey to the designer and constructor point by point requirements as to how a facility is to be designed and constructed. The guidelines should not be a rhetorical essay, but instead should follow an industry standard format, preferably a CSI (Construction Specifications Institute) format. Look at it this way, the material specifications tell the designer and constructor what material to use; the guidelines should tell them how to assimilate and use the material specifications in applying them to Good Design Practice. Without these guidelines as part of any bid package or Request For Proposal package, the Owner is essentially leaving it up to the Engineer and/or Constructor to bring their own set of guidelines to the table. And this may or may not be a good thing. Leaving the full facilities delivery to the Engineer and Constructor depends a great deal on the qualifications of the Engineer and the Constructor, and whether or not consistency from plant to plant and project to project is an issue. If the Owner approaches a project with expectations as to how they would like their plant or facility designed and built then some preparation, on the Owners part, is in order. Preparation should include, not only material specifications as described earlier, but also the guidelines and narratives (yes, narratives) necessary to define the design and construction requirements. I mention the use of narratives here because it helps facilitate the understanding and convey the magnitude of the, in most cases, reams of specifications and guidelines necessary to build an industrial facility of any appreciable size. A narrative, in general, should explain in simple, straight-forward language, for each discipline, the numbering scheme used for the specifications and guidelines; association between the material specifications and the guidelines; an explanation as to why the project is governed by a particular Code or Codes; and a brief description of expectation. The narrative allows you to be more explanatory and descriptive than a formal point-by-point specification. It gives the bidder/Engineer a Readers Digest version of the stacks of specifications and guidelines they are expected to read through and assimilate within a matter of a few weeks How piping specifications are delivered to a project can have a significant impact on the project itself. There are, generally speaking, three scenarios in which project specifications and guidelines are delivered to a project: 1. In scenario 1 the Owner, or Customer, has developed, throughout their existence, a complete arsenal of specifications and guidelines. In the older, more established 6

2.

3.

petroleum refining and chemical companies you will see entire departments whose mission is to create, maintain and refine all of the specifications and guidelines necessary to execute a project. When a project is approved to go out for bid to an Engineer the necessary specifications and guidelines, along with the requisite drawings, are assembled, packaged and provided to the Engineer as bid documents, and beyond that as working documents in the design, engineering and construction efforts. In scenario 2 the Owner, or Customer, has some specifications and guidelines that have possibly not been updated for several years. These are provided to the Engineer with the understanding and stipulation that any errors or omissions in the documents should be addressed and corrected by the Engineer. These too would be used in the bid process as well as on the project itself. In scenario 3 the Owner, or Customer, brings no specifications or guidelines to the project table. Specification development becomes part of the overall project engineering effort.

Scenarios 1 and 3 are at opposite ends of the spectrum, but afford the best situation for both the Owner and Engineer/Constructor. By providing the Engineer and Constructor, as in scenario 1, with a full set of current specifications and well articulated guidelines, making the assumption that both the engineer and constructor are qualified for the level of work required, they can very effectively execute the design, engineering and construction for the project. Scenario 3 allows the Engineer and Constructor to bring their own game-plan to the project. This too is effective, due only to the fact that the learning curve is minimal. Most engineering firms will be prepared to execute a project with their own set of specifications and guidelines. This applies to qualified Constructors as well. The down-side of this is the project to project inconsistency in specifications and methodology when using different engineers and constructors. Scenario 2 is a worse case situation. Ineffective and outdated Owner specifications create confusion and inefficient iterations in both the bid process and the execution of a project. It additionally creates the greatest opportunity for conflicts between Owner documents and the Engineers documents. For Project Management, this translates into change orders at some point in a project. A guideline should explain to the engineering firm or constructor, in a concise, definitive manner, just what it is the Owner expects of them in executing the design and construction of a facility. By actively and methodically developing a set of guidelines an Owner/Customer does not

have to rely on an outside resource, such as an engineering firm or constructor, to hopefully provide them with the facility they require and hope to get. Developing guidelines to convey your companys requirements and expectations can be accomplished using one or both of the following two basic methods: 1. A formal point-by-point format that covers all necessary criteria that you, as the Owner, require on a proprietary basis, plus a listing and description of the necessary Code and cGMP requirements. A narrative, for each discipline, that allows the writer to expand and define, in a much more descriptive manner, the points that arent made clear enough, or readily apparent in the more formal format.

wall thickness, MAWP/MADP, design pressure/temperature, and charge accumulation. Flanges In the learning process, some designers (I include engineers as well) will gloss over some of the primary basics of design and go directly to the bottom line information they need. Case in point: In Part I, of this series of articles, we discussed ASME flanges and their Classifications. Most designers are familiar with ASME flange Classifications such as 150, 300, 400, etc. And even though verbally stating 150 pound flange (we discussed the origin of this term in Part I) rolls off the tongue much easier and is still an industry accepted term; Class 150 is the proper terminology and designation. What you may not know is that the Class designation is a factor in the calculation for determining the rated working pressure of a flange. That calculation is:

2.

The guideline itself can be structured based on one of the CSI formats. The format examples provided by CSI give a company sufficient flexibility in writing guidelines, or specifications for that matter, to allow the document to conform to their own particular brand of requirements and nuances. It also lends a degree of intra-industry conformity to the guidelines and specifications, providing a degree of familiarity to the engineers and constructors that will have to adhere to them. Design Elements In the first paragraph of this segment of the article, Piping Design, I described the act of designing piping systems for a facility as bringing a number of technical components together to make the pipe conform to a specific set of requirements, within a prescribed area. Thats pretty simplistic, and does not really convey the magnitude of the experience, technical background or the imagination required to execute such a task. Experience is the essential component here. And that is simply because, aside from whatever innate ability a good designer might possess, the knowledge required is not taught through formal education, but is instead learned by being involved in the process of hands-on design over a period of time accompanied by ongoing learning. Ongoing learning can be in the form of organized classes, a mentor and/or any other means available to help learn and understand the physical requirements and restraints of the various systems you will be designing and industries you will be serving. Since we do not have enough space here to cover all of the design elements I would like to, I will key in on a few topics that generally find their way to me for clarification. And this doesnt even scratch the surface. We will discuss flanges, pipe internal surface finish, weld seam factor, pipe 7

PT
Where: Pc = PT Pr S1
1

Pr S1 / 8750

Pc

(eq. 1)

= = =

ceiling pressure, psig, as specified in ASME B16.5, para. D3 at temperature T Rated working pressure, psig,for the specified material at temperature T Pressure rating class index, psi1 (e.g., Pr = 300 psi for Class 300) Selected stress, psi, for the specified material at temperature T. See ASME B16.5, paras. D2.2, D2.3 and D2.4.

This definition of Pr does not apply to Class 150. See ASME B16.5, paras. D2.2, D2.3 and D2.4.

Pipe Internal Surface Finish Internal surface roughness is a topic that is specific to the pharmaceutical, bio-pharmaceutical and semiconductor industries, but can also be an issue in the chemical industry. Quantifying and specifying a maximum surface roughness for internal pipe wall for use in, what is referred to as direct impact fluid services, is a necessity in the above mentioned industries. Direct impact piping systems are those systems that carry product or carry a fluid service that ultimately comes in contact with product. The need for a relatively smooth internal pipe wall is predicated on three primary issues as it relates to the industries mentioned above. Those issues are: 1. Cleanability/Drainability, 2. The ability to hinder the growth (we dont yet have the ability to control it) of biofilm and to enhance the ability to remove it once it does appear, and 3. To reduce, to a microscopic level, crevices in which

microscopic particles can reside and at some point dislodge and get carried along in the fluid stream to damage product. Regarding the first point, cleanability and drainability are associative in this context. Meaning that, in order for a system to be fully cleanable it has to be designed and laid out in a manner that will eliminate any pockets and provide enough slope to eliminate any residual liquid (drainable). Not only is this residual liquid, or hold-up, a contaminant, from both a bacterial standpoint and as a cross batch contaminant, it can also be costly due to the high cost of some drug products. Along those lines, the ASME-BPE Standard provides criteria for minimum slope, maximum deadleg, gasket intrusion, gasket concavity, and many other criteria for design of cleanable and drainable hygienic piping systems. Regarding the second point, biofilm (Fig. 1) is defined as a bacterial population composed of cells which are firmly attached as microcolonies to a solid surface. A paper titled, Microbial Biofilms are they a problem in the Pharmaceutical Industry?, was delivered at an ASME-BPE symposium in Cork, Ireland, June 2004 by Frank Riedewald, a Senior Process Engineer with Lockwood-Greene IDC Ltd. In it he explains the results of testing that was performed to determine the relative association between the formation of biofilm, pipe wall surface finish and pipe wall surface cleanability.

The accepted max surface finish in the pharmaceutical and bio-pharmaceutical industries is 25Ra in (0.6 m). In the semiconductor industry you might typically see surface finishes in the 7Ra in to 15Ra in, particularly in gas delivery systems. While the pharmaceutical industry is concerned with bacterial growth and cross contamination, the semiconductor industry is concerned more with particulate damage to product, on the microscopic level. This pertains to point three above.

Fig. 2 Biofilm Attachment vs Surface Roughness (Courtesy of Mr. Riedewald) Pipe Weld Seam Factor Part I, of this series of articles, mentioned the fact that the weld seam in longitudinally welded pipe is a factor in the pipe wall pressure design thickness calculation. In ASME B31.3 there are two pipe wall thicknesses to calculate for. One is pressure design thickness (t) and the other is minimum required thickness (tm). There are two equations for finding pressure design thickness (t) for straight pipe under internal pressure. One is where t < D/6. This calculation (eq. 2) is based on internal pressure, actual (not nominal) OD of the pipe, stress value of the material at design temperature, joint efficiency factor, and the coefficient Y [a factor used to adjust internal pressure (P) for a nominal material at temperature]. The other calculation used is that in which t D/6. This calculation (eq. 3) is based on the above listed criteria except for the OD and uses instead ID of the pipe, and the sum of all mechanical allowances. The two equations look like this: Where t < D/6:
t PD 2( SE PY )

Fig. 1 Biofilm magnified 2000X (Courtesy of Mr. Riedewald) One of the many interesting factors that came from the studies mentioned in this paper is the fact that the internal surface of the pipe wall can actually be too smooth. Referring to the graph in Fig. 2, results of the studies in the above mentioned paper indicate that the surface finish range best suited to reduce biofilm adherence to the internal pipe wall surface is from 0.4Ra m to 1.Ra m (15.7Ra in to 58.8Ra in). What this implies is that while we currently do not have the means to prevent the onset of biofilm on the internal walls of hygienic or semiconductor piping systems we can facilitate its removal in the cleaning process by specifying the proper surface finish of the internal pipe walls. 8

(eq. 2)

Where t D/6:

P(d 2c) 2[ SE P(1 Y )]

(eq. 3)

Where: t = tm = c =

D d P S E Y

= = = = = =

Pressure design thickness Minimum required thickness, including mechanical, corrosion, and erosion allowances Sum of the mechanical allowances (thread or groove depth) plus corrosion and erosion allowances. For threaded components, the nominal thread depth (dimension h of ASME B1.20.1, or equivalent) shall apply. For machined surfaces or grooves where the tolerance is not specified, the tolerance shall be assumed to be 0.02 in. (0.5 mm) in addition to the specified depth of the cut. Actual pipe OD Pipe ID Internal design gage pressure Stress value for material from ASME B31.3 Table A-1, at design temperature Quality factor, or joint efficiency factor Coefficient from ASME B31.3 Table 304.1.1, valid for t < D/6.

Throughout design the vessels intended maximum pressure is referred to as its design pressure. All calculations are based on specified material and component tolerances along with fabrication specifics, meaning types and sizes of welds, reinforcement, etc. Not until after the vessel is fabricated can the engineer know what the actual material thickness is, the type and size of each weld, thickness of each nozzle neck, etc. Only when all of the factual data of construction is accumulated and entered into vessel engineering programs can the MAWP be determined. This value, once determined, then replaces the design pressure, and is calculated based on the installed configuration of the vessel. That is, mounted vertically or horizontally; mounted on legs; or mounted on lugs. The difference between the design pressure and the MAWP is that the engineer will design to the design pressure, but the final MAWP is the limiting pressure of the vessel, which may exceed the design pressure; it can never be less than the design pressure. In applying this to piping we will first calculate the burst pressure of the pipe and then determine the MAWP, or, as was mentioned earlier, a term more closely related to piping, the Maximum Allowable Design Pressure (MADP). There are three equations generally used in calculating burst pressure for pipe. They are: The Barlow formula;

The minimum required thickness (tm) is simple enough:

tm

(eq. 4)

PBA

2 TF S T D

(eq. 5)

To determine wall thickness for pipe under external pressure conditions refer to the Boiler and Pressure Vessel Code (BPVC) Section VIII, Division 1, UG-28 through UG30 and ASME B31.3, Para. 304.1.3. Keep in mind that for seamless pipe E will be removed from equations eq. 2 & eq. 3. Taking a page from the BPVC we will go through a few brief steps to determine Maximum Allowable Working Pressure (MAWP) for straight pipe. But let me begin by saying that MAWP is not a B31.3 expression, it comes from the BPVC. We will instead transpose this term to MADP (Maximum Allowable Design Pressure), which is also not a B31.3 term, but more closely relates to piping. When a vessel goes into design it is assigned a coincidental design pressure and temperature. These are the maximum conditions the vessel is expected to experience while in service, and what the vessel engineers will design to. The material, material thickness, welds, nozzles, flanges, etc. are all designed predicated on this predetermined design criteria.

The Boardman formula:

PBO

2 TF S T D (0.8 T )

(eq. 6)

The Lame` formula:

PL
Where: PBA PBO PL D d TF ST 9

ST

(D 2 d 2 ) (D 2 d 2 )

(eq. 7)

= = = = = = =

Burst Pressure, psig (Barlow Formula) Burst Pressure, psig (Boardman Formula) Burst Pressure, psig (Lame` Formula) Actual pipe OD, inches Pipe ID, inches Wall thickness, inches, minus factory tolerance Minimum tensile strength, psi, from

Sf M

= =

B31.3 Table A-1 Safety factor, a factor of 3 or 4 is applied to burst pressure to determine MADP Maximum Allowable Design Pressure (MADP)

System Operating Temperature: The temperature at which a fluid service is expected to normally operate at. System Design Temperature: Unless extenuating process conditions dictate otherwise, the design temperature, for operating temperatures between 32F and 750F, shall be equal to the maximum anticipated operating temperature, plus 25F rounded off to the next higher 5. Applying a sort of philosophy created by the above definitions is somewhat straight forward for utility services such as steam, water, non-reactive chemicals, etc. However, that part of the above definitions for design conditions that provide the caveat, extenuating process conditions implies a slightly different set of rules for process systems. Extenuating process conditions can mean increased pressure and temperature, beyond that defined above, due to chemical reaction, loss of temperature control in heat transfer, etc. Charge Accumulation of Lined Pipe Clarification Internal and external charge accumulation, known as static electricity, or more technically known, as triboelectric charge accumulation, is the result of charge generation unable to dissipate. If a charge generated in a flowing fluid is allowed to dissipate to ground, as it does in grounded metallic pipe, then there is no problem. However, if a charge cannot dissipate and is allowed to accumulate, it now becomes a problem by potentially becoming strong enough to create an Electrostatic Discharge (ESD). With regard to thermoplastic lined pipe there are two forms of this to be considered: External Charge Accumulation (ECA) and Internal Charge Accumulation (ICA). External Charge Accumulation ECA is a concern with lined pipe due to the possibility of not achieving spool-to-spool continuity during installation due, in large part, to improved paint primer on flanges. To explain the loss of spool-to-spool continuity: this lack of integral continuity is, as mentioned above, the result of the prime paint coat that is applied by the manufacture. When pipe spools, lined or un-lined, are joined by flanges using non-metallic gaskets the only thing that completes the Spool-to-spool continuity is the bolting. The improved paint primer on lined pipe flanges makes this more difficult to achieve because normal bolt tightening doesnt guarantee metal-to-metal contact between the nut and the flange. Pipe generally does not come with a prime coat of paint, however lined pipe does. Since flange bolts are used to complete continuity from spool to spool the installer has to make certain, when installing lined pipe, that the bolts, at least one of the bolts, has penetrated the primer and made contact with bare metal. This was achieved in the past by 10

Using the results from any one of the above equations we can then solve for MADP as follows:

P* * Sf

(eq. 8)

** = BA, BO, or L subscript Design Pressure and Temperature The ASME B31.3 definition for Design Pressure and Design Temperature is stated as two separate definitions. I will integrate them into one by stating: The design pressure and temperature of each component in a piping system shall be not less than the most severe condition of coincident internal or external pressure and temperature (minimum or maximum) expected during service. It goes on to state: The most severe condition is that which results in the greatest required component thickness and the highest component rating. How do you determine these values and where do you apply them? Well cover the where first. What we did earlier in determining pipe wall thickness was based on design conditions, in which P = Internal Design Gage pressure and S = Stress value at design temperature. Design conditions are also used to determine component rating and as a basis for determining leak test pressure, which we will get into in the final article of this series. There is no published standard, or real industry consensus on how to determine design conditions. It basically comes down to an Owners or engineers experience. What I will provide here is a resultant philosophy developed from many sources along with my own experiences. In understanding what constitutes design conditions we first of all need to define them. Following is some accepted terminology and their definitions: System Operating Pressure: The pressure at which a fluid service is expected to normally operate at. System Design Pressure: Unless extenuating process conditions dictate otherwise, the design pressure is the pressure at the most severe coincident of internal or external pressure and temperature (minimum or maximum) expected during service, plus the greater of 30 psi or 10%.

using star washers on at least one flange bolt while assuming possible bare metal contact with the other bolts allowing the washers, as they were tightened, to scrape away the prime coat so that contact was made with the bare metal of the flange. With improved prime coat material this is no longer a guarantee. If continuity from spool to spool is not achieved any charge generation resulting from an internal or an external source cannot readily dissipate to ground. The voltage in triboelectric charge generation will build until it is strong enough to jump to the closest grounded object creating an undesired spark of electricity in doing this (Electrostatic Discharge). Internal Charge Accumulation ICA, with regard to pipe, is unique to thermoplastic lined pipe and solid thermoplastic pipe. Without being impregnated with a conductive material, thermoplastics are not good conductors of electricity. PTFE (Polytetrafluoroethylene), as an example, used as a pipe liner, has a high (>1016 Ohms/Square), resistivity factor. This is a relatively high resistance to conductivity. This means that any charge created internally to the pipe cannot readily be conducted away to ground by way of the PTFE liner. Instead the charge will be allowed to build until it exceeds its total dielectric strength and burns a pinhole in the liner to the internal metal wall of the casement pipe. It isnt charge generation itself that is the problem, its the charge accumulation. When the rate of charge generation is greater than the rate of charge relaxation (the ability of material to conduct away the generated charge), charge accumulation occurs. The dielectric strength of PTFE is 450 to 500 volts/mil. This indicates that for every 0.001 of PTFE liner 450 volts of triboelectric charge will be required to penetrate the liner. For a 2 pipeline with a 0.130 thick liner this translates into 58500 volts of triboelectric charge to burn through the liner thickness. When the liner is penetrated by an accumulated charge two additional problems (time bombs) are created: 1. Corrosive fluid (a major user of lined pipe) is now in contact with and corroding the metal pipe wall and at some point, depending on rate of corrosion, will fail locally causing fluid to leak to the environment, and 2. The initial charge that burned through the liner is now charging the outer metal pipe, which, if continuity has not been achieved for the outer pipe, a spark of triboelectric charge is, at some point, going to jump to ground causing a spark. Corrective Action External Charge Generation The simplest method to ensure continuity is to sand away any primer on the back side of each flange to ensure good metal-to-metal contact between nut and flange. Aside from that or the use of a conductive prime paint, the current 11

ready-made solution to the external continuity problem is the addition of stud bolts located in close proximity to flanges on both pipe spools and fittings (see Fig. 3). These studs can be applied at the factory or in the field. At each flange joint a grounding strap (jumper) is then affixed to a stud on one spool with a nut, extended over the flange joint and attached to a stud on the connecting spool completing continuity throughout the chain of connecting spools and fittings.

Figure 3 Grounding Lug Location Another method of creating continuity at flange joints, while being less obtrusive and more integral, is described as follows and represented in Fig. 4: Referring to Fig. 4, flanges would be purchased predrilled and tapped in the center of the outer edge of the flange between the backside of the flange and the face side of the flange. The drilled and tapped hole in each flange will need to be centered between boltholes so that they line up after the flange bolts are installed. The tapped hole is 1/4 dia. x 1/2 deep. After a flange set is installed and fully bolted the Continuity Plate (Fig. 4) can be installed using two 1/4 x1/2 long hex head screws and two lock washers. The Continuity Plate has two 0.312 slotted boltholes allowing for misalignment and movement. The entire continuity plate assembly is relatively simple to install, unobtrusive and establishes integral contact with the pipeline.

Figure 4 Continuity Flange Plate Internal Charge Generation One of the first options in preventing Internal Charge Accumulation is by minimizing charge generation. This can be done by adjusting the flow velocity relative to the liquids conductivity. To minimize design impact, cost and even schedule impact on a project this needs to be evaluated early in the project due to the possibility of a change in line size. To retard charge generation by reducing flow velocities British Standard (BS) suggests the following as represented in Table 1 per BS 5958: TABLE 1 - RECOMMENDED VELOCITIES BS 5958 Recommended Flow Liquid Conductivity Velocity >1000 pS/m No restriction 50 1000 pS/m Less than 7 m/s Less than 50 pS/m Less than 1 m/s
pS/m (picosiemens/meter)

Figure 5 Conductive Orifice Plate Assembly

Figure 6 Conductivity Orifice Plate Assembly Section Conclusion and Recommendations It is difficult to pre-determine what fluid services and systems will be candidates for charge accumulation prevention and Electrostatic Discharge protection. The simplest and most conservative answer to that is to assume that all fluid services in lined pipe systems are susceptible. In saying that, we then have to declare that a companys pipe specifications need to reflect a global resolution that will affect all installations. With regard to External Charge Accumulation, the recommendation for future installations with the least impact would be to specify pipe with no prime coat or at least no primer on the flanges, or a prime coat using a conductive paint. The un-primed pipe would be primed prior to installation with care given to primer touchup on flanges after installation by the installing contractor or their sub. This would better ensure spool-to-spool external continuity. For existing installations either the studs or the continuity plate installation would work. It can also be suggested that the continuity plates can be tacked on to one flange rather than drilling and tapping both flanges. For dissipating internal charge generation the orifice plate, as shown in Figures 5 & 6, is the only recommendation. PIPE FABRICATION Entering this part of the article on fabrication does not mean that we leave engineering behind. Indeed, the majority, if not all, fabricators (referring to the fabricators that are qualified for heavy industrial work) will have an engineering staff. 12

If velocity reduction is not an option, or if further safeguards against charge accumulation are warranted then a mechanical solution to provide a path to ground for Internal Charge Generation might be called for. One method for conducting charge accumulation from the interior of the pipe to ground is indicated in Figures 5 & 6. What is shown is an orifice plate made of conductive (static dissipative) material that is compatible with the fluid service. The orifice itself is off center to the OD of the plate and the pipeline itself. With the shallow portion of the ID at the invert of the pipe it allows the piping to drain in horizontal runs. The tab portion of the plate extends beyond the flange OD. On the tab is a bolthole for attaching the modified Continuity Flange Plate. The plate is designed to come in contact with the interior surface of the liner wall as well as protrude into the flowing fluid providing a conduit for internally generated charge. Continuity is achieved by attaching the plate to the flange OD that is in contact with the piping, which is, in turn, grounded through equipment.

Hygienic Clamp Joint As a project moves from the design phase into the construction phase anyone with a modicum of project experience can acknowledge the fact that there will most certainly be conflicts, errors and omissions, no matter how diligent one thinks they are during design. This is inherent in the methodology of todays design/engineering process. There are methods and approaches to design in which this expected result can be minimized. Its actually a retrospective concept, but we will save that for a future article. The preparation for such errors and omissions is always prudent. If, on the other hand, the assumption is made that the Issued for Construction design drawings will facilitate fabrication and installation with minimal problems, then you can expect to compound whatever problems do occur because you werent prepared to handle them. The greatest asset a project manager can have is the ability to learn from past experience and the talent to put into practice what they have learned. Fabrication Pipe fabrication, in this context, is the construction of piping systems by forming and assembling pipe and components with the use of flanged, threaded, clamped, grooved, crimped and welded joints. In Article I we discussed the flange joint, we will discuss the others here. There are various factors, or considerations, that prompt the decision as to which type of connection to use in the assembly of a piping system. To start with, any mechanical joint is considered a potential leak point and should be minimized. Also, the decision as to which type of joint should be specified comes down to accessibility requirements, installation requirements and joint integrity. Using that as our premise we can continue to discuss the various joining methods. Threaded Joints Pipe thread, designated as NPT (National Pipe Taper) under ASME B1.20.1, is the type of thread used in joining pipe. This is a tapered thread that, with sealant, allows the threads to form a leak-tight seal by jamming them together as the joint is tightened. I described the threaded flange joint in Article I. Those same criteria apply also to threaded fittings, in which the benefits of the threaded joint is both in cost savings and in eliminating the need for welding. In this regard, to paraphrase Article I, threaded components are sometimes used in high-pressure service in which the operating temperature is ambient. They are not suitable where high temperatures, cyclic conditions or bending stresses can be potential concerns. The clamped joint, as mentioned in Article I, refers to the sanitary or hygienic clamp. As you can see in Fig. 7 there are issues with this type clamp.

Figure 7 Hygienic Clamp Joint (Courtesy Rubber Fab Technologies Group) Represented in Fig. 7 are three installed conditions of the hygienic joint, minus the clamp. Joint A represents a clamp connection that has been over tightened causing the gasket to intrude into the ID of the tubing. This creates a damming effect, preventing the system from completely draining. In joint B the clamp wasnt tightened enough and left a recess at the gasket area. This creates a pocket where residue can accumulate and cleanability becomes an issue. Joint C represents a joint in which the proper torque was applied to the clamp leaving the ID of the gasket flush with the ID of the tubing. The clamp C representation is the result that we want to achieve with the hygienic clamp. The problem is that this is very difficult to control on a repeatable basis. Even when the gasket and ferrules are initially lined up with proper assembly and torque on the joint, some gasket materials have a tendency to creep (creep relaxation), or cold flow. Creep relaxation is defined as: A transient stress-strain condition in which strain increases concurrently with the decay of stress. More simply put, it is the loss of tightness in a gasket, measurable by torque loss. Cold Flow is defined as: Permanent and continual deformation of a material that occurs as a result of prolonged compression or extension at or near room temperature. There have been a number of both gasket and fitting manufacturers that have been investing a great deal of research in attempting to resolve this issue with the clamp joint. Some of the solutions regarding fittings were 13

addressed in Article I. Additionally, gasket manufacturers, and others have been working on acceptable gasket materials that have reduced creep relaxation factors, as well as compression controlled gasket designs. When mentioning acceptable gasket material in the previous paragraph, what I am referring to is a gasket that is not only compatible with the hygienic fluid service, but also meets certain FDA requirements. Those requirements include Gasket material that complies with USP Biological Reactivity Test #87 & 88 Class VI for Plastics and FDA CFR Title 21 Part 177. Grooved Joint The grooved joint (Fig. 8), from simply a static internal pressure containment standpoint, is as good as or, in some cases superior to the ASME Class 150 flange joint. In the smaller sizes, 1 through 4 the working pressure limit will be equal to that of a Class 300, carbon steel, ASTM A105, ASME B16.5 flange. Its main weakness is in its allowable bending and torsional stress at the coupling. This can be alleviated with proper support. Because of this design characteristic the manufacturers of grooved joint systems have focused their efforts and created a niche in the fire protection and utility fluid service requirements, with the exception of steam and steam condensate. This type of joint is comparatively easy to install and enhances that fact in areas that would require a fire card for welding. Since no welding is required modifications can be made while operation continues. Some contractors choose to couple at every joint and fitting, while others choose to selectively locate couplings, much as you would selective locate a flange joint in a system. Its a decision that should be made based on the particular requirements or preference of a project or facility.

The pressed joint is actually a system that uses thin wall pipe, up through 2 NPT, to enable the joining of pipe and fittings with the use of a compression tool. Welding is not required and threading is only necessary when required for instrument or equipment connection.

Figure 9 Pressed Joint (Courtesy Victaulic) These types of systems are available from various manufacturers in carbon steel, 316 and 304 stainless steel and copper. Because of the thin wall pipe corrosion allowance becomes a big consideration with carbon steel. While the static internal pressure rating of these systems is comparable to an ASME Class 150 flange joint there are additional fluid service and installation characteristics that need to be considered. With axial and torsional loading being the weak spot in these systems they are not practical where water hammer is a potential, such as in steam condensate. The axial load consideration carries over to supporting the pipe as well. Ensure that vertical runs of this pipe are supported properly from beneath. Do not allow joints in vertical runs to be under tension. They must be supported properly from the base of the vertical run. Welded Joint The welded joint is by far the most integrated and secure joint you can have. When done properly it is as strong as the pipe itself. The key to a welds integrity lies in the craftsmanship of the welder or welding operator, the performance qualification of the welder or welding operator, and the weld procedure specification. Before I go further I want to explain the difference between the terms welder and welding operator. A welder is, as you might have guessed, someone who welds. To be more precise, it is someone who welds by hand, or manually. A welding operator is someone who operates an automatic welding machine. The ends of the pipe still have to be prepared and aligned, and the automatic welding machine has to be programmed.

Figure 8 Grooved Pipe & Coupling (Courtesy Victaulic) Pressed Joint

The advantage of machine welding is apparent in doing production welds. This is shop welding in which there is a quantity of welds to be made on the same material type, wall thickness and nominal pipe size. Once the machine is set up 14

for a run of typical pipe like this it is very efficient and consistent in its weld quality. This is another topic that could easily stand alone as an article, but we wont do that here. Instead we will focus on some of the primary types of welding used with pipe. Those types include: 1. 2. 3. 4. GMAW (Gas Metal Arc Welding) or MIG GTAW (Gas Tungsten Arc Welding) or TIG SMAW (Shielded Metal Arc Welding) or MMA or Stick Welding FCAW (Flux Cored Automatic welding)

material it is welding it will provide excellent welds on a consistent basis. Provided, that is, that the chemistry of the base material is within allowable ranges.

GMAW: Most often referred to as MIG, Metal Inert Gas welding, GMAW (Gas Metal Arc Welding) can be an automatic or semi-automatic welding process. It is a process by which a shielding gas and a continuous, consumable wire electrode is fed through the same gun (Fig. 10). The shielding gas is an inert or semi-inert gas such as argon or CO2 that protects the weld area from atmospheric gases, which can detrimentally affect the weld area. There are four commonly used methods of metal transfer used in GMAW. They are: 1. globular, 2. short-circuiting, 3. spray, and 4. pulsed-spray With the use of a shielding gas the GMAW process is better used indoors or in an area protected from the wind. If the shielding gas is disturbed the weld area can be affected.

Figure 11 GTAW (TIG) Welding (Courtesy The Welding Institute) A wide differential in sulfur content between the two components being joined can cause the weld to drift into the high sulfur side. This can cause welds to be rejected due to lack of full penetration. SMAW: Also referred to as MMA, Manual Metal Arc welding, or just simply Stick welding, SMAW (Shielded Metal Arc Welding) is the most common form of welding used. It is a manual form of welding that uses a consumable electrode, which is coated with a flux (Fig. 12). As the weld is being made the flux breaks down to form a shielding gas that protects the weld from the atmosphere. The SMAW welding process is versatile and simple, which allows it to be the most common weld done today.

Figure 10 GMAW (MIG) Welding (Courtesy The Welding Institute) GTAW: Most often referred to as TIG, Tungsten Inert Gas welding, GTAW (Gas Tungsten Arc Welding) can be automatic or manual. It uses a nonconsumable tungsten electrode to make the weld (Fig. 11), which can be done with filler metal or without filler metal (autogenous). The TIG process is more exacting, but is more complex and slower than MIG welding. In Article 1 I mentioned the use of orbital welding for hygienic tube welding. Orbital welding uses the GTAW method. Once the orbital welder is programmed for the 15

Figure 12 SMAW (Stick) Welding (Courtesy The Welding Institute) FCAW: Flux Cored Arc Welding is a semiautomatic or automatic welding process. It is similar to MIG welding, but the continuously fed consumable wire has a flux core. The flux provides the shielding gas that protects the weld area from the atmosphere during welding. Welding Pipe The majority of welds you will see in pipe fabrication will be full penetration circumferential buttwelds, fillet

welds or a combination of the two. The circumferential buttwelds are the welds used to weld two pipe ends together or other components with buttweld ends. Fillet welds are used at socketweld joints and at slip-on flanges. Welds in which a combination of the buttweld and fillet weld would be used would be at a stub-in joint or a joint similar to that. A stub-in joint (not to be confused with a stub-end) is a connection in which the end of a pipe is welded to the longitudinal run of another pipe (Fig. 13). Depending on what the design conditions are this can be a reinforced connection or an unreinforced connection. The branch connection can be at 90 or less from the longitudinal pipe run.

11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

Gage Calibration certifications Weld Continuity Report WPSs (Weld Procedure Specifications) PQRs (Procedure qualification Record) Weld Coupon log Weld Maps Slope Maps Weld Logs Leak Test Reports Inspection reports Passivation Records Detail mechanical layouts technical specifications for components As-Built Isometrics Original IFC isometrics Documentation recording any changes from IFC to As-Build isometrics

Figure 13 Sample Stub-In Connections (Courtesy ASME B31.3) Hygienic Fabrication and Documentation Hygienic and semiconductor pipe fabrication uses automatic autogenous welding in the form of orbital welding. This, as explained in Article I, is a weld without the use of filler metal. It uses the orbital welding TIG process. In some cases hand welding is required, but this is kept to a minimum, and will generally require pre-approval. When fabricating pipe for hygienic services it will be necessary to comply with, not only a specific method of welding, but also an extensive amount of documentation. As mentioned in Article I, developing and maintaining the required documentation for hygienic pipe fabrication and installation can add an additional 30% to 40% to the piping cost of a project. The documentation needed, from the fabrication effort for validation, may include, but is not limited to: 1. 2. Incoming Material Examination Reports Material Certification a. MTRs b. Certification of Compliance 3. Weld Gas Certification 4. Signature Logs 5. WPQs (Welder & Welding Operator Performance Qualification) 6. Welder & Welding Operator Inspection Summary 7. Mechanical and electropolishing procedures 8. Examiner Qualification 9. Inspector Qualification 10. Welder Qualification Summary 16

The above listed documentation, which closely parallels the list in ASME-BPE, is that which is generally required to move an installed hygienic system through validation, commissioning and qualification (C & Q). And this isnt all thats required. There is additional supporting documentation such as P&IDs, procedural documents, etc. that are also required. Depending on the size and type of a project it can be a massive undertaking. If not properly set up and orchestrated it can become a logistical nightmare. What you do not want to do is discover during C&Q that you are missing a portion of the required documentation. Resurrecting this information is labor intensive and can delay a projects turn-over significantly. I cannot stress it strongly enough just how imperative it is that all necessary documentation be identified up front. It needs to be procured throughout the process and assimilated in a turnover package in a manner that makes it relatively easy to locate needed information while also allowing the information to be cross indexed and traceable within the TO package. The term validation is a broad, generalized, selfdefining term that includes the act of commissioning and qualification. Commissioning and qualification, while they go hand in hand, are two activities that are essentially distinct within themselves. For this article I will go no further with the topic of Validation, Commissioning and Qualification. This is a topic that I will touch on again in Article III. Future Articles The third and final article in this series, titled Piping Design Part III Installation, Cleaning, Testing and Verification, will wrap up the series by discussing the four title points. Acknowledgement:

I wish to thank Earl Lamson, Senior Project Manager with Eli Lilly and Company, for being kind enough in taking time out of a busy schedule to read through the draft of this second article. Earl has a remarkable set of project and engineering skills that set him apart from many I have worked with. That and the fact that I value his opinion are the reasons I asked him to review this article. About the author: W. M. (Bill) Huitt has been involved in industrial piping design, engineering and construction since 1965. Positions have included design engineer, piping design instructor, project engineer, project supervisor, piping department supervisor, engineering manager and president of W. M. Huitt Co. a piping consulting firm founded in 1987. His experience covers both the engineering and construction fields and crosses industrial lines to include petroleum refining, chemical, petrochemical, pharmaceutical, pulp & paper, nuclear power, biofuel, and coal gasification. He has written numerous specifications, guidelines, papers, and magazine articles on the topic of pipe design and engineering. Bill is a member of ISPE (International Society of Pharmaceutical Engineers), CSI (Construction Specifications Institute) and ASME (American Society of Mechanical Engineers). He is a member of three ASME-BPE subcommittees, several Task Groups, an API Task Group, and sets on two corporate specification review boards. He can be reached at: W. M. Huitt Co. P O Box 31154 St. Louis, MO 63131-0154 (314)966-8919 wmhuitt@aol.com www.wmhuitt.com

17

As published in the September and October 2007 issues of Chemical Engineering Magazine

Piping Design Part 3: Installation, Cleaning, Testing & Verifification

Efficiency, quality and safety are the imperatives that are factored in when considering field fabrication, but dont forget cost.
W. M. (Bill) Huitt W. M. Huitt Co. As the title implies this article will discuss the Installation, Cleaning, Testing and, to a lesser degree, Validation of piping systems. I say to a lesser degree with Validation because Validation is a very complex, often proprietary and exceedingly difficult-to-define topic. Rather than delve into it in great detail as part of a multi-topic article I will attempt to simply provide some understanding as to its function and need. PIPE INSTALLATION But first things first, the installation of pipe follows its fabrication and is very frequently a part of it. The installation of pipe can be accomplished in the following four primary ways, or combinations thereof: 1. 2. 3. 4. Field fabricate and install, Shop fabricate and field erected, Skid fabrication, assembly & installation, and Modular construction at an on-site field fabrication area and then erected. A number of factors will dictate whether or not it is feasible to field fabricate: The size and type of the project, pipe size and material, the facility itself, weather conditions, availability of qualified personnel, existing building operations, cleanliness requirements, time available to do the work, etc. Efficiency, quality and safety are the imperatives that are factored in when considering field fabrication. And before you think I missed it, cost is the fallout of those factors. Logistically speaking, if all pipe could be fabricated on-site in a safe and efficient manner, maintaining quality while doing so, it would make sense to do it in that manner. However, before making that final decision, lets look at some of the pros and cons of field fabrication: Pros: 1. Only raw material (pipe, fittings, valves, etc.) need to be shipped to the site location. This is much easier to handle and store than multi-plane configurations of pre-fabricated pipe. 2. No time-consuming need to carefully crib, tiedown and chock pre-fabricated *spool pieces for transport to the job site. 3. Reduced risk of damage to spool pieces. 4. More efficient opportunity to fab around unexpected obstacles (structural steel, duct, cable tray, etc.) 5. Fabricate-as-you-install reduces the rework risk assumed when pre-fabricating spools, or the cost

I would like to assure you that I am not going to diverge off into fabrication again since we discussed it, although somewhat briefly, in Article II. I am including fabrication in this article simply because fabrication is such an integral part of pipe installation. FIELD FABRICATE AND INSTALL Field fabrication and installation is just what it implies. The pipe is fabricated on site either in place or in segments

6.

related to field verification prior to shop fabrication. The field routing installation of pipe through an array of insufficiently documented locations of existing pipe and equipment, on a retrofit project, is quite frequently more effective than attempting to pre-fabricate pipe based on dimensional assumptions.

One is a Field Weld (FW) and the other is a Field Closure Weld (FCW). The FW indicates a joint in which the end of a pipe segment is prepared for the installer to set in place and weld to its connecting joint without additional modification in the field. This means that the length of pipe that is joined to another in the field is cut precisely to length and the end prepared in the shop for welding. The FCW provides the installer with an additional length of pipe, usually 4 to 6 longer than what is indicated on the design drawings, to allow for field adjustment. What has to be considered, and what prompts the need for a FCW, is the actual, as-installed, location of both the fixed equipment that the pipe assemblies may connect to and the actual installed location of the pipe assembly itself. Odds are that all equipment and piping will not be installed exactly where indicated on design drawings. The dimensional location of the equipment items given on design drawings is not a finite location, they are merely intended locations, as are drawings for building steel, pipe supports and others. What factors into the installation of shop fabricated pipe is the actual location of the equipment nozzle it will be connecting to in relation to the pipes installed location. In connecting to equipment there is a build-up, or stackup, of tolerances that will effectively place the actual, or final, location of the nozzle at some point in the xyz geometry of three-dimensional space, other than where the design drawing indicates. The tolerance stack-up comes from the following: 1. 2. 3. 4. Manufacturing tolerances in material forming, nozzle location, and vessel support location. The actual set-in-place location of the vessel. Load cell installation (when applicable). The actual set-in-place pipe run-up location.

*Spool pieces are the pre-fabricated sections of pipe that are fabricated and numbered in the fab shop then shipped to the job site for installation.

Cons: 1. Weather is arguably the biggest deterrent. If the facility under construction is not enclosed then protection from the elements will have to be provided. 2. When welding has to be done in conditions that are not environmentally controlled then preheating will be required if the ambient temperature (not the metal surface temperature) is 0 F or below. 3. In a new facility, as opposed to having to route piping through an array of poorly located existing pipe and equipment, field fabrication of buttwelded pipe is not as efficient and cost effective as shop fabrication. 4. Concerns about safety and efficiency when working in a facility while it is in operation in advance of a turnaround or to begin advance work on a plant expansion. Generally speaking, threaded, socketweld, grooved, and other proprietary type joints that do not require buttwelding are field fabricated and installed. Buttwelding of small, 1 1/2 NPS and less, are very often field fabricated and installed because of the added risk of damage during transport, in pre-fabricated form, from the shop to the site. SHOP FABRICATE AND INSTALL Shop fabrication is, generally speaking, any pipe, fittings and components that are assembled by welding into spool assemblies at the fabricators facility. The spools are then labeled with an identifier and transported to the job site for installation. Each spool piece needs its own identifier marked on the piece itself in some fashion that will make it easy to know where its destination is in the facility and/or where it belongs in a multi-spool system of pipe. This will allow the installer to efficiently stage the piece and ready it for installation. As part of the process of developing spool sections field-welded joints need to be designated. These are welded joints that connect the pre-fabricated spools. In doing this the designer or fabricator will identify two different types of field-welded joints.

In order to allow for these inevitable deviations between the drawing dimensions used to fabricate the vessel, set the vessel, and install the pipe assembly, and the actual installed location of the connecting points, a field closure piece, or two, will be required for that final adjustment. The field closure piece is a designated section of the pipe assembly in which a field weld has been indicated. The section with the field closure weld would be the length required to agree with that indicated on the design drawing, plus an additional 4 to 6 (more or less depending on fabricators comfort level with the equipment locations). What this does is allow the field to make the final determination in the adjustments when connecting to fixed equipment. SKID (SUPER SKID) FABRICATION

A skid is a pre-packaged assembly that may contain all or some of the following that make up an operating system: vessels, rotating equipment, piping, automation components, operator interfaces, instrumentation, gages, electrical panels, wiring and connectors, framework, supports, in-line piping components, and insulation. A single process or utility system may fit onto one skid or, depending on size restraints, may comprise multiple skids. After fabrication of a skid is complete it will typically go through Factory Acceptance Testing (FAT) at the fabricators facility. The skid is then shipped to the job site where it is installed in its final location. After installation it would typically go through a follow-up Site Acceptance Test (SAT), including additional hydro-testing. This is basically a system shake-down to determine that everything is intact, and that those things that did not remain intact during transport are discovered and repaired. Logistics and the necessary skill-set required for the installation, connection and start-up of a particular skid package will dictate to what extent the skid fabricator will be involved after it is shipped to the job site. MODULAR CONSTRUCTION The term module or modular construction is quite often, in this context, interchanged with the term skid fabrication. A module can refer to pre-fabricated units that actually form the structure of a facility as each is installed. Or, the units may be smaller sub-assemblies that, when combined, make up a complete process or utility system. Modules too consist of all or some of the following: vessels, rotating equipment, piping, automation components, HVAC, instrumentation, electrical wiring and connectors, framework, walls, architectural components, lighting, supports, in-line piping components, and insulation. This, as an example, allows a complete locker room module to be placed and connected to a complete water treatment module. The smaller sub-assembly modules, in many cases, are interchanged with the term skid. It saves on misperception when a company defines these terms, both for internal discussion and for the purpose of making it clear to outside contractors, as to what is meant when using the term module. INSTALL APPROACH Now that we have a general idea of the four primary approaches to piping installations how do we decide which is the best method, or combination of methods, to use for a particular project? But there is one major caveat I would like to address before launching into this subject. Each project is individualized with its own particular set of decision drivers with regard to a selected execution 3

approach. There are no hard and fast rules for determining a best approach at job execution. It requires experienced personnel assigning values to the various aspects of project execution, overlaying a timeline, and then assessing logistics. Sounds simple, but is in actuality can be a very complex process. What I am attempting to say here is, that the following is a guideline and not hard and fast rules. There are simply too many project variables and complexities to allow it. In approaching this decision keep in mind that the method of installation needs to be weighed against a contractors preferred methodology. In saying that I am not implying that the contractors preferred methodology should drive your decision on how to execute a job. On the contrary, once you determine how the job needs to be executed you then look to only those contractors whose preferred methodology agrees with your project execution plans. Some contractors prefer to do most, if not all fabrication in the shop, others prefer to set up at the job-site, while others are flexible enough to utilize the best of both methods. The three main criteria, efficiency, quality and safety indicated earlier under Field Fabricate and Install, would apply here as well. Using those three elements as a basis for making our determination let us look at some common variables: 1. Environment a. Controlled environment b. Open to the elements Industry a. Pharmaceutical b. Biopharmaceutical c. Semiconductor d. Food & Dairy e. Petroleum refining f. Bulk chemical g. Pulp & paper h. Off-Shore i. Pipeline j. Power generation Type of project a. Retrofit b. Fast track approach c. New (Grassroots/Greenfield) project d. Clean-build e. Single level f. Multi-level g. Room repetition Range of pipe material and sizes a. Small percentage of alloy pipe b. Large percentage of alloy pipe c. Large % of large pipe sizes

2.

3.

4.

5.

d. Large % of small pipe sizes e. Mix of small and large pipe sizes Location a. Close to metropolitan area b. Remote location c. Country with limited resources

of a facility, to be microbial and particulate free, as stipulated by the design. There can be no debris, organic or inorganic, remaining after construction in accessible or inaccessible spaces of the facility. Of particular concern with the pharmaceutical, biopharm and food 7 dairy is food waste and hidden moisture. Food waste can entice and support rodents and insects, and hidden moisture can propagate mold, which can eventually become airborne. If not discovered until the facility is in operation the impact, upon discovery, can potentially be devastating to production. Such contamination can be discovered in one of two ways. Discovery at the source, possibly behind a wall or some other out-of-the-way place, means that not only does current production have to cease, but product will have to be analyzed for possible contamination. Once found it hen has to be remediated. The other method of discovery comes from the continuous testing and validation of the product stream. If a contaminant is discovered in the product the production line is stopped and the problem then becomes an investigation in to finding the source of the contamination. The clean-build philosophy therefore dictates more stringent and strict requirements for controlling and inspecting for debris on an ongoing basis throughout construction and start-up. It will be necessary, on a clean-build site, to follow some rather simple rules: 1. Smoking or smokeless tobacco products of any kind are not allowed on the site property, 2. Off site break and lunch areas, no food or drink, other than water, allowed on the site premises, 3. Do not begin installing pipe, duct or equipment until, at the very least, a roof is installed, 4. After roof and walls are installed ensure that there is no standing water remaining in the facility, 5. Prior to and during the construction of hollow walls, such as those framed and dry-walled, ensure on a daily basis that there is no moisture or debris in the wall cavity, 6. Duct work delivered to the job site shall have the ends covered with a plastic sheet material, which shall remain on the ends until connected in place, 7. Fabricated pipe delivered to the job site shall have the ends covered in a suitable fashion with suitable material, and shall remain on the ends until connected in place, 8. During and after flushing and testing of pipelines all water spills shall be controlled to the extent possible and shall be cleaned up after flushing and testing or at the end of the work day,

Environment The environment is only a factor when work has to be done in an open-air structure or other outdoor installation (tank farm, pipeline, pipe rack or yard piping, etc.). Working in an open air structure will require protection from the elements (rain, snow, wind, cold, etc.). There may additionally be a requirement to work in elevated areas on scaffolding and otherwise. All of this can have a potential impact on safety and efficiency. Pipe rack installation consists mainly of straight runs of pipe, and will not necessarily have a requirement or need for pre-fabrication. That is, unless it is pre-fabricated as modular skid units. Depending on the project it could be cost effective on an overall strategic basis to modularize the pipe rack, steel and all. The big advantage to shop fabrication is the controlled environment in which its done. This includes the Quality Control aspect, better equipment (generally speaking), a routine methodology of how a piece of work progresses through the shop, and better control, through a developed routine, of required documentation. Industry I know this is generalizing, but we can group the various industries into clean/indoor build and nonclean/outdoor build. There are exceptions to this, but under clean/indoor build we can list the following; Clean/Indoor build a. Pharmaceutical b. Biopharmaceutical c. Semiconductor d. Food & Dairy Under non-clean/outdoor build we can list the following; Non-Clean/Outdoor Build a. Petroleum refining b. Bulk chemical c. Pulp & paper d. Off-Shore e. Pipeline f. Power generation The clean build philosophy comes from the need to construct certain facilities with a more stringent control on construction debris. Those industries listed above under Clean/Indoor Build often require a facility, at least a portion 4

Type of Project While the type of project is not the main influence in determining how you approach the execution of a project it does play a key role. It will help drive the decision as to how the piping should be fabricated and installed. As an example, if the project is a retrofit it will require much of the pipe, regardless of size and joint connection, to be field fabricated and installed. This is due simply to the fact that the effort and cost necessary to verify the location of all existing pipe, equipment, walls, columns, duct, etc. in a somewhat precise manner, would not be very practical. You would be better served by field verifying the approximate location of the above items with existing drawings, for planning and logistic purposes, then shop or field fabricate, verify and install as you go. A fast track project, one that has a compressed schedule, will require parallel activities where possible. Whereas shop and skid fabrication would be utilized as much as possible simply to expend more man-hours over a shorter time period while attempting to maintain efficiency. Even though there may be added cost to this approach. This approach is time driven and not budgetary driven. A new grassroots facility still requires routing verification as you go, but certainly not the much more involved need to locate previously installed obstructions as needs to be done when working with an existing facility. If the project is a clean-build project (typical for the pharmaceutical, biopharmaceutical, semiconductor and food & dairy industries) inside an environmentally controlled area it will be more practical to shop fabricate or utilize skid or modular fabrication for most, if not all of the piping. This will reduce the number of personnel and the amount of fabrication debris in the facility, and provide better control for keeping it out of the pipe itself. With personnel you could have food wrappers, drink cans and bottles, food waste, and clothing items. Fabrication debris could include metal filings, cutting oil, pieces of pipe, weld rod and weld wire remnants, etc. If the project is not a clean-build, but is still inside an environmentally controlled facility the same logic does not necessarily apply. The decision to shop fab and install or to field fab and install becomes one based on efficiency rather than how best to maintain a clean area. And thats not to say that if it doesnt qualify as a clean-build project then the construction debris can just be allowed to pile up. There is still safety and efficiency to be concerned with on any project and a clean job site is a major part of that. Maintaining a clean job site is an integral component of good project execution.

Keeping personnel and equipment to a minimum at the job site is not an absolute, but is one of the key considerations to the efficiency of pipe installation. Following that logic most of the buttwelded pipe should be shop fabricated. A couple of things to consider, when determining which buttwelded pipe to shop fabricate, is size and material. Range of Pipe Material and Sizes Shop fabricated spools need to be transported to the job site. This requires handling. Handling and transporting small diameter pipe and/or thin-wall tubing spools creates the potential for damage to those spools. If you are shop fabricating everything and the distance from shop to site is simply across town the risk to damaging small diameter pipe spools is a great deal less than if they have to be shipped half way across the US, Europe or Asia. Or even across an ocean. In transporting spools over long distances, unless there is a great deal of thought and care given to cribbing the load of spools, it may not be beneficial to transport buttwelded pipe spools NPS 1 and less. It may be more practical to fabricate these sizes on site, unless you are fabricating hygienic or semi-conductor piping. These types of systems require a great deal more control and a cleaner fabrication. Meaning that pipe fabrication will require a clean shop area on-site, or the pipe will need to be fabricated at an off-site, better controlled shop facility. A practical rule of thumb in determining what to shop fab and what to field fab follows in Table 3-1: Table 3-1 Shop and Field Fabrication Size (in) Material Joint Shop/Field 1 Pipe 1, 2, 3, 6 Field 1 Pipe 4&5 Shop 2 Pipe 3&6 Field 2 Pipe 4&5 Shop 1 Tubing 5 Field 1 Tubing 5 Shop (a, b) 1 Tubing 5 Shop
Joint Type: 1 = Socketweld 2 = Threaded 3 = Grooved Fully (Grooved fittings and pipe ends.) 4 = Grooved Partially (Shop-welded spools with grooved ends.) 5 = Buttweld 6 = Flanged Lined or unlined Pipe Notes: a. Hygienic tubing b. Special cribbing and support for transport

The above Table 3-1 is a general methodology. Dictates of the project and a contractors SOP will determine how best to define what gets shop fabricated and what gets field fabricated.

Petroleum refining and bulk chemical projects are generally open air projects in which field fabrication and installation of pipe is exposed to the elements. While a clean build is not a requirement on these types of projects efficiency and, above all, safety is, as it is on any type project. Because of this, it would make sense to utilize shop fabrication as much as possible. Fabricating pipe spools under better controlled shop conditions will provide improved efficiency and safer per hour working conditions over what you will generally find in the field. This translates into fewer accidents. Referring back to Table 3-1, with respect to the potential for damage during transport, pipe sizes NPS 2 to 3 and larger ship much better than smaller pipe sizes. Particularly when working with thin-wall tubing. This is a consideration when determining what to shop fab and what to field fab. Location Job site location is one of the key markers in determining shop or field fabrication. In many cases building a facility in a remote location will be a driver for utilizing a disproportionate amount of skid or module fabrication. Disproportionate in the sense that project management may look at modularizing the entire job, rather than mobilize the staffing and facilities needed to fab and install on or near the job site. This would constitute a larger amount of modularization over what might normally be expected for the same type project in a more metropolitan region, or an area with reasonable access to needed resources. To expand on that thought; it was pointed out to me by Earl Lamson, Senior Project Manager with Eli Lilly and Company, an observation I fully agree with, that project resources, even in metropolitan areas, are quite frequently siloed around a specific industry. In certain regions of the US for example, you may discover that there are an abundance of craftsman available when building a refinery, but that same region may have difficulty, from a trained and experienced personnel perspective, in supporting the construction of a semiconductor facility. Consequently when building a pharmaceutical facility in another region you may find a sufficient population of trained and experienced craftsman for that industry, but may not find that resource adequate when building a chemical plant. Building a project in a remote location requires the project team to rethink the job-as-usual methodology. From a logistics standpoint mobilization of personnel and material become a major factor in determining the overall execution of such a project. Project planning is a big component in project execution, but is more so when attempting to build in

remote areas. And this doesnt even touch on the security aspect. Nowadays, when constructing in any number of remote areas, security is a real concern that requires real consideration and real resolution. Reduced on-site staffing is a good counter-measure in reducing risk to personnel when building in remote or even non-remote third-world areas. PIPE SYSTEM CLEANING While there are requirements in ASME for leak testing cleaning requirements do not exist. In ASTM A 380 & 967 you will find Standards on cleaning, descaling and passivation, but nothing in ASTM on simply flushing and general cleaning. Defining the requirements for the internal cleaning of piping systems falls within the responsibilities of the Owner. The term cleaning, in this context, is a catch-all term that also includes flushing, chemical cleaning, and passivation. So before we go further let me provide some definition for these terms as they apply in this context. I say, as they apply in this context, because these terms are somewhat flexible in their meaning, depending on source and context, and could be used to describe activities other than what is intended in this dialog. Definitions Cleaning: A process by which water, solvents, acids or proprietary cleaning solutions are flushed through a piping system to remove contaminants such as cutting oils, metal filings, weld spatter, dirt, and other unwanted debris. Flushing: A process by which water, air or an inert gas is forced through a piping system either in preparation for chemical cleaning or as the only cleaning process. Flushing can be accomplished by using dynamic pressure head or released static pressure head, as in a fill-and-dump procedure. Blow-down can be considered as flushing with a gas. Passivation: A process by which a chemical solution, usually with a base of nitric, phosphoric, citric acid or other mild oxidant, is used to promote or accelerate the formation of a thin (25 to 50 Angstroms) protective oxide layer (a passive layer) on the internal surface of pipe, fittings and equipment. In stainless steels, the most commonly used alloy at present, it removes any free iron from the pipe surface to form a chromium-rich oxide layer to protect the metal surface from aggressive liquids such as high purity waters.
Note: Cleaning and Flushing can be interchanged when the process only requires water, air or an inert gas to meet the required level of cleanliness. When the term cleaning is used in this context it may infer what is defined as flushing.

Cleaning and Testing 6

With regard to cleaning and leak testing, and which to do first, there are drivers for both and different schools of thought on the overall process. Each contractor will have their preference. It is in the Owners best interest to determine their preference or be at risk in just leaving it to the contractor. In either case you should have a line of thought on the process, if for no other reason than to be able to understand what it is the contractor is proposing to do. At the very least, in advance of leak testing, perform either a basic flush of a *test circuit, or perform an internal visual examination as the pipe is installed. A walk-down of the test circuit should be done just prior to filling the system with any liquid. The last thing you want to happen is to discover too late that a joint wasnt fully connected or an inline component was taken out of the pipeline. In a facility that is not a clean-build it can simply be a mess that has to be cleaned up. In a clean-build facility an incident such as this can potentially be costly and time-consuming to remediate.
Note: *refer to the following section on Leak Testing

One other thing I would like to mention before we go on. Since we are discussing new pipe installation we will not include steam-out cleaning or pipeline pigging. These are cleaning procedures that are used on in-service piping to clean the fluid service residue build-up from interior pipe walls after a period of use. Before subjecting the system to an internal test pressure the piping should first be walked down to make certain, as mentioned earlier, that there are no missing or loose components. The system is then flushed with water or air to make sure that there are no obstacles in the piping. Over the years we have discovered in installed piping systems everything from soda cans to shop towels, work gloves, nuts & bolts, weld rod, Styrofoam cups, candy wrappers, and other miscellaneous debris including dirt and rocks. After this initial flush, which could also be the only flush and cleaning required, the system is ready for chemical cleaning or to leak test. In large systems it may be beneficial to leak test smaller test circuits and then perform a final cleaning once the entire system is installed and tested. This would include a final completed system leak test that would test all of the joints that connect the test circuits. That is, unless these joints were tested as the assembly progressed. If it is decided, on large systems, to leak test smaller segments, or test circuits as they are installed, prior to flushing the entire system, the piping needs to be examined internally as it is installed. This is to prevent any large debris items, as listed above, from remaining in the piping during the test. Now that we have touched on those generalities lets take a look at each of the cleaning Categories listed in Table 3-2 and see how to apply them. Cleaning Category C-1 is simply a flush with water, air or inert gas. The one non-manual assist that water requires in order for it to clean the inside of a piping system is velocity. But what velocity is necessary? The main concept behind flushing a pipeline is to dislodge and remove suspected debris. In order to dislodge, suspend and remove this unwanted material in the piping system it is necessary that water or air be forced through the piping system at a velocity sufficient to suspend the heaviest suspected particles and move them along the pipeline. The velocity required to suspend the particles and move them along the pipeline for removal is dependent upon their size and weight, and the flush medium. Metal filings, arguably the heaviest particles normally found in newly fabricated pipe, will have a terminal mid-range settling velocity, in water, of approximately 10 feet per second. Therefore, a flushing velocity of approximately 10 feet per second should be achieved during the flush. (This does not apply to acid cleaning.) 7

Before getting into further specifics of this discussion we need to define some general cleaning and testing procedures and assign them some easy to use indicators. In this way it will be much easier to discuss the various processes. We can then work through some general scenarios and see which sequencing works best. Following is a list of cleaning requirements: Table 3-2 General Cleaning Scenarios Category Description C-1 Flush only (water, air or inert gas) C-2 Flush, clean with cleaning solution, flush C-3 Clean with cleaning solution, flush C-4 Flush, clean, passivate, flush Following is a list of leak testing requirements: Table 3-3 General Leak Testing Scenarios Category Description T-1 Initial service leak test T-2 Hydrostatic leak test T-3 Pneumatic leak test T-4 Sensitive leak test T-5 Alternative leak test While the cleaning descriptions are self explanatory the leak testing descriptions may not be. Please refer to the following section on Leak Testing to find clarification of the terms used in Table 3-3.

The following Table 3-4 indicates the rate of flow required to achieve approximately 10 feet per second of velocity through various sizes and schedules of pipe.
Table 3-4 Rate of Flushing Liquid Needed to Maintain Approximately 10 FPS Velocity (GPM) Pipe Sch. 5s 40 80 12 10 7 20 16 13 Pipe Sizes (inches) 1 34 27 22 1 77 64 55 2 123 105 92 3 272 230 4 460 397

circulation of a cleaning solution, which is then followed by a final flush of water. Cleaning solutions are, in many cases, proprietary detergent or acid-based solutions each blended for specific uses. Detergent-based solutions are generally used for removing dirt, cutting oils and grease. Acid-based solutions are used to remove the same contaminants as the detergentbase plus weld discoloration and residue. The acid based solution also passivates the pipe wall. As defined earlier, passivation provides a protective oxide barrier against corrosion. The acids used in some cleaning solutions for ferrous and copper materials leave behind a passivated interior pipe surface as a result of the cleaning process. In utility water services such as tower water, chilled water, etc., this barrier against corrosion is maintained with corrosion inhibitors that are injected into the fluid stream on an ongoing basis. And keep in mind that when I talk about passivated surfaces this is a natural occurrence with metals in an oxygen environment. The acid merely initiates and speeds up the process. When using stainless alloys, usually 316L, in hygienic water services such as Water For Injection (WFI), Purified Water, Deionized (DI) Water and in some cases Soft Water, passivation is a final intended step in the preparation for service of these pipelines. Passivation is also a periodic ongoing preventative maintenance procedure. To explain: High purity water is very corrosive and attacks any free iron found on the surface of stainless pipe. Free iron has a tendency to come out of solution when material is cold worked, as in bending or forming pipe without the benefit of heat. It also occurs with the threading of alloy bolts, which are solution annealed (heat treated) after threading. Passivation removes this free iron while also accelerating, in the presence of O 2, the oxidation rate of the stainless steel providing a chromium rich oxide corrosion barrier as defined above. Over time (and this is one hypothetical thought on the subject), this very thin corrosion barrier tends to get depleted or worn off, particularly at high impingement areas of the piping system (elbows, tees, pump casings, etc.). Once the passive layer wears through any free iron exposed to the high purity water will oxidize, or rust. This will show up as surface rouge. Rouging is an unwanted surface discoloration which is periodically removed by means of a derouging process. This is an operational, as-needed chemical cleaning process that will remove all or most of the rouge and also re-passivate the internal pipe surface.

Purging a piping system clear of debris with air requires a velocity of approximately 25 feet per second. The following Table 3-5 indicates the rate of air flow required to achieve approximately 25 feet per second of velocity through various sizes and schedules of pipe.
Table 3-5 Rate of Air Flow to Maintain Approx 25 FPS Velocity (SCFS) Pipe Sch Press 15 psig Press 50 psig 5s 40 80 5s 40 80 Pipe Sizes (inches) 0.14 0.11 0.08 0.30 0.23 0.18 0.23 0.19 0.15 0.51 0.41 0.33 1 0.39 0.30 0.25 0.84 0.66 0.55 1 0.86 0.71 0.62 1.88 1.56 1.35 2 1.39 1.18 1.04 3.02 2.56 2.26 3 3.06 2.59 2.32 6.67 5.65 5.05 4 5.17 4.47 4.03 11.3 9.73 8.79

One thing you might notice is that the size range only extends to 4 NPS for both the liquid flush and for the air or gas blow-down. The reason for that is the volume of liquid or gas required to achieve the necessary velocity through the larger pipe sizes is quite significant. As an example a 6 NPS pipeline would require approximately 900 to 1000 GPM, depending on wall thickness of the pipe, to achieve a velocity of 10 FPS. This gets a little cumbersome and costly. That is unless you have pumps or compressors in place that can achieve the necessary flow rate. The alternative for liquid flushing the larger pipe sizes other than using source line pressure or a pump is to perform a fill-and-dump. In this process the pipe system is completely filled with liquid and then drained through a full line size, quick opening valve. In doing this there has to be enough static head to generate sufficient force and velocity to achieve essentially the same result as the pumped or line pressure liquid. Cleaning Category C-2 is a three-step process by which the piping system is initially flushed out with a liquid to remove most of the loose debris. This is followed by the 8

Discussions and research on the topic of rouging continues. This is a subject that has more questions than answers at the present time. Currently the ASME-BPE is looking into this issue. One of the questions to be answered is whether or not rouge is actually detrimental to product streams. Cleaning Category C-3 is a two-step cleaning process that uses a detergent or acid based solution to clean the pipe interior of any unwanted residue or debris. This is then followed by a final flush of water. Cleaning Category C-4 is a three or four-step process generally used in hygienic service piping. In most cases, simply due to the clean fabrication approach used in hygienic pipe fabrication, only a water flush with Deionized (DI) quality water or better would be necessary for cleaning followed by passivation of the piping system, then a final flush of water. There are variations to each of these primary cleaning functions and it would be in an Owners best interest to define these requirements, by fluid service, in advance of the work to be done. LEAK TESTING Pressure testing is a misnomer that is quite often used when referring to leak testing of piping systems. And as long as all parties understand what is meant by that, then thats fine. However, in a true sense a pressure test is a test you perform on a relief valve to test its set point pressure. The intent, when pressure testing a relief valve, is not to check for leaks, but to test the pressure set point of the valve by gradually adding pressure to the relief valve until it lifts the valve off of the seat. A leak test, on the other hand, is performed to check the sealing integrity of a piping system by applying internal pressure to a pre-determined limit, based on design conditions, then checking joints and component seals for leaks. It is not intended that the MAWP of a piping system be verified or validated. Before discussing the various types of leak tests and leak test procedures I would like to briefly talk about controlling and tracking this activity. Cleaning and testing, like many aspects of a project, should be a controlled process. Meaning, there should be a formal method of documenting and tracking this activity as the Contractor proceeds through the leak testing process.. In documenting the leak testing activity there are certain forms that will be needed. They consist of the following: 1. A dedicated set of P&IDs to identify the limits and number the test circuits; 9

2. 3.

4.

A form to record components that were either installed or removed prior to testing; A checklist form for field supervision to ensure that each step of the test process is accomplished; and Leak test data forms

The two sets of documents, from those listed above, that need to be retained are the P&IDs (#1) and the Leak Test Data Forms (#4). The other two sets of forms are procedural checklists. The Leak Test Data forms should contain key data such as: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. Test circuit number P&ID number(s) Date of test Project name and/or number Location within facility Line number(s) Design pressure Test pressure Test fluid Test fluid temperature Time (military) recorded test begins Pressure at start of test Time (military) recorded test ends Pressure at end of test Total elapsed time of test Total pressure differential (plus or minus) from beginning to end of test period 17. Comment section (indicate if leaks were found and system was repaired and retested or if system passed) 18. Signatures & dates Also make certain that the testing contractor has current calibration logs of their test instruments, such as pressure gages. To continue with the leak testing, ASME B31.3 defines five primary leak tests as follows: Initial Service Leak Test: This applies only to those fluid services meeting the criteria as defined under ASME B31.3 Category D fluid service. This includes fluids in which the following apply: (1) the fluid handled is nonflammable, nontoxic, and not damaging to human tissue; (2) the design gage pressure does not exceed 1035 kPA (150 psi); and (3) the design temperature is from -29C (-20F) through 186C (366F). The Initial Service leak test is a process by which the test fluid is the fluid that is to be used in the intended piping system at operating pressure and temperature. It is accomplished by connecting to the fluid source with a

valved connection and then gradually opening the source valve and filling the system. In liquid systems air is purged during the fill cycle through high point vents. A rolling examination of all joints is continually performed during the fill cycle and for a period of time after the system is completely filled and is under line pressure. In a situation in which the distribution of the pipeline that is being tested has distribution on multiple floors of a facility there will be pressure differentials between the floors due to static head differences. This will occur in operation and is acceptable under initial service test conditions. The test pressure achieved for initial service testing pressure is what it is. Meaning that what you achieve in the test is what it will be in operation. The only difference is that the flowing fluid during operation will incur an amount of pressure drop that will not be present during the static test. Hydrostatic leak test: This is the most commonly used leak test and is performed by using a liquid, normally water, and in some cases with additives to prevent freezing, under a calculated pressure.

energy in the pressurized gas. And for that reason alone it should be used very selectively. When pneumatic testing is performed it must be done under a strictly controlled procedure with on-site supervision in addition to coordination with all other crafts and personnel in the test area.

P T P T P T

1.1P 1.4 P 1.2Pto1.5P

(eq. 3)

(eq. 4)

(eq. 5)

The test pressure for pneumatic leak testing under B31.3 is calculated using eq. 3, for B31.9 it is calculated using eq. 4, and for B31.1 it is calculated using eq. 5. One misconception I need to address here with pneumatic leak testing is in its procedure, as described in B1.3. There is a misconception that the test pressure should be maintained while the joints are examined. This is not correct. As B31.3 explains, pressure is increased gradually until the test pressure is reached. At that point the test pressure is held until piping strains equalize throughout the system. (eq. 1.1) After allowing a sufficient amount of time for piping strains to equalize the pressure is then reduced to the design pressure (refer to article II for design pressure). While holding design pressure all joints are examined for leaks. It is not required that the examination take place while holding test pressure. There is more to the entire procedure that I did not include here. Please refer to B31.3 or B31.1 for full details on pneumatic leak testing. Sensitive leak test: This leak test is performed when there is a higher than normal potential for fluid leakage, such as for hydrogen. I also recommend its use when a fluid is classified as a Category M fluid service. B31.1 refers to this test as Mass-Spectrometer and Halide Testing. In B31.3 the process for sensitive leak testing is as follows: The test shall be in accordance with the gas and bubble test method specified in the BPV Code, Section V, Article 10, or by another method demonstrated to have equal sensitivity. Sensitivity of the test shall be not less than 10 -3 atm-ml/sec under test conditions. (a) The test pressure shall be at least the lesser of 105kPa (15 psi) gage, or 25% [of] the design pressure.

PT

1.5PST S

(eq. 1)

Where: PT = Test Pressure, psi P = Internal design gage pressure, psig ST = Stress value at test temperature, psi (see B31.3 Table A-1) S = Stress value at design temperature, psi (see B31.3 Table A-1 Eq. 1 represents the equation for that calculated pressure. However, as long as the metal temperature of ST remains below the temperature at which the allowable stress value for ST begins to diminish and the allowable stress value of S and ST are equal then ST and S cancel each other leaving the simpler eq. 2:

P T

1.5P

(eq. 2)

Unlike initial service testing, pressure variations due to static head differences in elevation have to be accommodated in hydrostatic testing. What I mean by that is the calculated test pressure is the minimum pressure required for the system. When hydrostatically testing a multi-floor system the minimum calculated test pressure shall be realized at the highest point. This is not stated, but is inferred in B31.3. Pneumatic leak test: This test is performed using air or a preferred inert gas. This is a relatively easy test to perform simply from a preparation and cleanup standpoint. However, this test has a hazardous potential because of the stored 10

(b) The pressure shall be gradually increased until a gage pressure the lesser of one-half the test pressure or 170 kPa (25 psi) gage is attained, at which time a preliminary check shall be made. Then the pressure shall be gradually increased in steps until the test pressure is reached, the pressure being held long enough at each step to equalize piping strains. In testing fluid services that are extremely difficult to seal against, or fluid services classified as a Category M fluid service I would suggest the following in preparation for the process described under B31.3: prior to performing the sensitive leak test perform a low pressure (15 psig) test with air or an inert gas using the bubble test method. Check every mechanical joint for leakage. After completing the preliminary low pressure pneumatic test, purge all of the gas from the system using helium. Once the system is thoroughly purged, and contains no less than 98% helium, continue using helium to perform the sensitive leak test with a helium mass spectrometer. Helium is the trace gas used in this process and has a molecule that is close to the size of the hydrogen molecule making it nearly as difficult to seal against as hydrogen without the volatility. Test each mechanical joint using the mass-spectrometer to determine leak rate, if any. Alternative leak test: In lieu of performing an actual leak test, in which internal pressure is used, the alternative leak test takes the examination and flexibility analysis approach. This test is conducted only when it is determined that hydrostatic or pneumatic testing would be detrimental to the piping system and/or the fluid intended for the piping system, an inherent risk to personnel, or impractical to achieve. As an alternative to testing with internal pressure it is acceptable to qualify a system through examination and flexibility analysis. The process calls for the examination of all groove welds, and includes longitudinal welds used in the manufacture of pipe and fittings that have not been previously tested hydrostatically or pneumatically. It requires a 100% radiograph or ultrasonic examination of those welds. Where applicable, the sensitive leak test shall be used on any untested mechanical joints. This Alternative leak test also requires a flexibility analysis as applicable. Very briefly, a flexibility analysis verifies, on a theoretical basis, that an installed piping system is within the allowable stress range of the material and components under design conditions if a system: (a) duplicates or replaces without significant change, a system operating with a successful service record; (b) can be judged adequate by comparison with previously analyzed systems; and (c) is of uniform size, has no more than two points of fixation, no 11

intermediate restraints, and falls within the limitations of empirical equation (eq. 6).

Dy (L U )2
Where:

K1

(eq. 6)

D = outside diameter of pipe. in. (mm) y = resultant of total displacement strains to be absorbed by piping system, in. (mm) L = developed length of piping between anchors, in. (mm) U = anchor distance, straight line between anchors, ft. (m) K1 = 208,000 SA/Ea, (mm/m)2 = 30 SA/Ea, (in./ft.)2 SA = allowable displacement stress range per equation (1a) of ASME B31.3, ksi (MPa) Ea = reference modulus of elasticity at 70F (21C), ksi (MPa) One example in which an alternative leak test might be used is in making a branch tie-in to an existing, in-service line using a saddle with an o-let branch fitting with a weld neck flange welded to that and a valve mounted to the flange. Within temperature limitations, the fillet weld used to weld the saddle to the existing pipe can be examined using the dye penetrant or magnetic particle method. The circumferential butt or groove weld used in welding the weld neck and the o-let fitting together should be radiographically or ultrasonically examined. And the flange joint connecting the valve should have the torque of each bolt checked after visually ensuring correct type and placement of the gasket. There are circumstances, regarding the tie-in scenario we just discussed for alternative leak testing, in which a hydrostatic or pneumatic test can be used. It depends on what the fluid service is in the existing pipeline. If it is a fluid service that can be considered a Category D fluid service then it is quite possible that a hydrostatic or pneumatic leak test can be performed on the described tie-in. By capping the valve with a blind flange modified to include a test rig of valves, nipples and hose connectors, you can perform a leak test rather than an alternative leak test. As mentioned this does depend on the existing service fluid. If the existing fluid service is steam or a cryogenic fluid then you might want to consider the alternative leak test. Cleaning and Leak Testing Procedures As you can see by equations eq. 1 through eq. 5 above, the leak test pressure, except for initial service testing, is based on design pressure and design temperature. In Article 2 we described design pressure and temperature. What we will do here is apply that understanding and describe a few general procedures for cleaning and testing.

As in all other project functions control and documentation is a key element in the cleaning and testing of piping systems. It does, however, need to be handled in a manner that is dictated by the type of project. Meaning that you dont want to bury yourself in unwarranted paperwork and place an unneeded burden on the contractor when it isnt necessary. Building a commercial or institutional type facility will not require the same level of documentation and stringent controls that an industrial type facility would require. But even within the industrial sector there are varying degrees of required testing and documentation. To begin with, documentation requirements in industry standards are simplistic and somewhat generalized, as is apparent in ASME B31.3, which states in Para. 345.2.7: Records shall be made of each piping system during the testing, including: (a) date of test (b) identification of piping system tested (c) test fluid (d) test pressure (e) certification of results by examiner These records need not be retained after completion of the test if a certification by the inspector that the piping has satisfactorily passed pressure testing as required by this Code is retained. ASME B31.3 goes on to state, in Para. 346.3: Unless otherwise specified by the engineering design, the following records shall be retained for at least 5 years after the record is generated for the project: (a) examination procedures; and (b) examination personnel qualifications. Standards, that cover such a broad array of industrial manufacturing, do not, as a rule, attempt to get too specific in some of their requirements. Beyond the essential requirements, such as those indicated above, the Owner, engineer or contractor has to assume responsibility and know-how for providing more specific and proprietary requirements for a particular project specific to the particular needs of the Owner. The following will help, to some extent, fill that gap. Cleaning Procedures This section will describe some fundamental cleaning procedures as they might appear in a specification or guideline, and this includes the leak test procedures that follow. This will give you some idea as to what you might consider developing for your own set of specifications. Assuming that if your company repeatedly executes projects you will have cleaning and testing guidelines, in some form, 12

prepared for your contractor. If not you may not get what you expect. Its better to give some forethought to these activities rather than be surprised at the results. Once a menu of these cleaning and testing procedures are developed, using pre-assigned symbols, much as those given in the following, they can then be specified in the line list with the respective fluid services as you require. In this manner there is no second guessing during construction. Each piping circuit is assigned a specific clean and test protocol in advance. Many pre-developed procedures I have seen over the years, those developed by Owners in particular, have been very simplistic, and typically out of date. This is an indicator to most contractors that the Owners Rep will most likely not attempt to enforce them. The contractor, in making that assumption, may simply ignore them and perform their own procedures. What your procedural guidelines should do is be explicit enough and current to the point where the contractors know that someone has given some thought as to how they want that work accomplished. Making it far more likely they (the contractors) will execute your procedure instead of theirs. It is certainly acceptable to accommodate suggestions to a procedure from a contractor when it doesnt compromise the intent of the Owners requirements and improves the efficiency of the contractor. If a submitted alternate procedure does not compromise the intent of the Owner it is recommended that they be accepted. This will allow the Owner to see if that efficiency is really there. With that in mind lets create a couple of general cleaning procedures. A general practice in the flushing and cleaning process, also indicated in leak testing, is the evacuation of air when using liquids. Always provide high point vents for evacuating air during the fill cycle and low point drains for clearing out all of the liquid when the process is complete. Using the same symbology indicated in Table 3-2 these cleaning procedures will be categorized as follows: Category C-1: Flush or Blow Down only (water, air or inert gas) C-1.1 These systems shall be flushed with the fluid that the system is intended for. There shall be no hydrostatic or pneumatic leak test. An Initial Service leak test will be performed. Refer to test Category T-1. a. Connect system to its permanent supply line. Include a permanent block valve at the supply line connection. All outlets shall have temporary hoses run to drain. Do not flush through coils, plates, strainer or filter elements.

b. Using supply line pressure, flush system through all outlets until water is clear and free of any debris at all outlet points. Flush a quantity of fluid through each branch not less than three times that contained in the system. Use Table 3-6 to estimate volume of liquid in the system. c. These systems are required only to undergo an Initial Service leak test. During the flushing procedure, and as the system is placed into service, all joints shall be checked for leaks. d. Any leaks discovered during the flushing process, or during the process of placing the system into service, will require the system to be drained and repaired. After which the process will start over with step 2.

If no debris is found the system is ready for leak testing. Category C2: Flush then clean with cleaning solution, followed by a neutralization rinse. Because of the thoroughness of the flush, clean and rinse process there should be no need to check for transient debris, only for neutralization. However, if circumstances dictate otherwise then a final check for debris may be warranted. C-2.1 These systems shall be pre-flushed with potable water, cleaned with (indicate cleaning agent) then a rinse/neutralization followed by leak testing with potable water. If it is determined that the system will be installed and tested progressively in segments, the sequence of cleaning and testing can be altered to follow the segmented installation. Thereby leak

Table 3-6 Volume of Water Per Lineal Foot of Pipe (gal.) Pipe Sizes (inches) Sch. 5s 20 40 80 1/2 .021 .016 .012 3/4 .035 .028 .023 1 .058 .045 .037 11/2 .129 .106 .093 2 .207 .176 .154 3 .455 .386 .345 4 .771 .664 .60 6 1.68 1.51 1.36 8 2.71 2.61 10 4.31 4.11 12 6.16 5.84 14 7.34 9.22 16 9.70 9.22 18 12.4 14.5 20 15.2 14.5 24 22.2

C-1.2 These systems shall be flushed clean with Potable Water. a. Connect a flush/test manifold at a designated inlet to the system, and a temporary hose or pipe on the designated outlet(s) of the system. b. Route temporary hose or pipe from potable water supply, approved by Owner, and connect to flush/test manifold. Route outlet hose or pipe to sewer, or as directed by Owner rep. Secure end of outlet. c. Using a Once through procedure (not a recirculation), and the rate of flow in Table 3-4, perform an initial flush through the system with a quantity of potable water not less than three times that contained in the system. Use Table 3-6 to estimate volume of liquid in the system. Discharge to sewer, or as directed by Owner rep. d. After the initial flush, insert a conical strainer into a spool piece located between the discharge of the piping system and the outlet hose. Perform a second flush with a volume of potable water not less than that contained in the system. e. After the second flush (step 4), pull the strainer and check for debris; if debris is found repeat step 3.

testing segments of a piping system as they are installed without cleaning. The entire system would then cleaned once installed and tested. a. Hook up flush/test manifold at a designated temporary inlet to the system between the circulating pump discharge and the system inlet. Install a temporary hose or pipe on the designated outlet(s) of the system. b. Route temporary hose or pipe from potable water supply, approved by Owner, and connect to flush/test manifold. Route outlet hose or pipe to sewer, or as directed by Owners Rep. c. Close valve between the circulating pump (if no valve included in the system design insert a lineblind or install a blind flange with a drain valve) discharge and flush/test rig. Open valve between flush/test manifold and piping system. d. Using the once through procedure (meaning the cleaning fluid is not re-circulated), and the rate of flow in Table 3-4, perform an initial flush through the system, by-passing the circulation pump, with a quantity of potable water equal to not less than three times that contained in the system. Use Table 3-6 to estimate volume of liquid in the system. (Note: During the water flush check the system for leaks. Verify no leaks prior to introducing chemical cleaning solution to the piping system.) 13

e. Discharge to sewer, or as directed by Owners Rep. f. After completing the initial flush, drain remaining water in the system. Or, retain water if cleaning chemicals will be added to the circulating water. g. Configure valves and hoses to circulate through pump. Connect head tank, or other source containing cleaning agent, to connection provided on circulation loop. h. Fill the system with the pre-measured (indicate preferred cleaning agent and mixing ratio or % by volume) and circulate through the system for 48 hours. To minimize corrosion, if anticipated, circulate cleaning agent at a low velocity rate prescribed by the cleaning agent manufacturer. i. Drain cleaning agent to sewer or containment, as directed by Owner. j. Reconnect as in step #1 for the once through flush/neutralization, and flush system with potable water using a quantity not less than three times that of the system volume. Since the (name cleaning agent) solution has a neutral pH the rinse water will have to be visually examined for clarity. Rinse until clear. The rinse must be started in as short as quickly after the cleaning cycle as possible. If cleaning residue is allowed to dry on the interior pipe wall, it will be more difficult to remove by simply flushing. The final rinse and neutralization must be accomplished before any possible residue has time to dry. k. Test pH for neutralization. Once neutralization is achieved proceed to step #12. l. Remove pump and temporary circulation loop then configure the system for leak testing. This may include removal of some components, insertion of line-blinds, installation of temporary spools pieces, etc. Those three examples should provide an idea as to the kind of dialog that needs to be created in providing guidance and direction to the contractor responsible for the work. And, as I stated earlier, these procedures, for the most part, are flexible enough to accommodate suggested modifications from the contractor. Leak Test Procedures As in the cleaning procedures we will keep this general, but provide enough specifics for you to develop leak testing

procedures that will suit your companys own particular needs. In Article 1 I stated the B31.3 definition for Category D fluid services. I then indicated that while Category D fluid services qualified for initial service leak testing there are caveats that should be considered. Again, this is a situation in which ASME provides some flexibility in testing by lowering the bar on requirements where there is reduced risk in failure. Provided, that if failure should occur the results would not cause catastrophic damage to property or irreparable harm to personnel. The Owners responsibility, for any fluid service selected for initial service leak testing lies in determining what fluid services to place into each of the fluid service Categories. Those Categories being: Normal, Category D, Category M, and High Pressure. Acids, caustics, volatile chemicals and petroleum products are usually easy to identify as those not qualifying as a Category D fluid service. Cooling tower water, chilled water, air, and nitrogen are all easy to identify as qualifyiers for Category D fluid services. The fluid services that fall within the acceptable Category D guidelines, but still have the potential for being hazardous to personnel are not so straight forward. Using water as an example, at ambient conditions water will simply make you wet if you get dripped or sprayed on. Once the temperature of water exceeds 140F (60C), by OSHA standards, it starts to become detrimental to personnel upon contact. At this point the range of human tolerance becomes a factor. However, as the temperature continues to elevate it eventually moves into a range that increasing becomes scalding upon human contact and human tolerance is no longer a factor because it is now hazardous and the decision is made for you. Before continuing I need to be clear on the above subject matter. The 140F temperature mentioned above is with respect to simply coming in contact with an object at that temperature. Brief contact at that temperature would not be detrimental. In various litigation related to scalding it has been determined that an approximate one-second exposure to 160F water will result in third degree burns. An approximate half-minute exposure to 130F water will result in third degree burns. And an approximate ten minute exposure to 120F water can result in third degree burns. With the maximum temperature limit of 366F (185.5C) for Category D fluid services what the Owner needs to consider here are three factors: within that range of 140F (60C), the temperature at which discomfort begins to set in, to 366F (185.5C), the upper limit of Category D fluids, what do we consider hazardous; what is the level of 14

opportunity for risk to personnel; and what is the level of assured integrity of the installation. What I mean by assured integrity is this: if there are procedures and protocols in place that require, validate and document third-party inspection of all pipe fabrication, installation and testing, then there is a high degree of assured integrity in the system. If some or all of these requirements are not in place then there is no assured integrity. All three of these factors: temperature, risk of contact, and assured integrity, have to be considered together to arrive at a reasonable determination for borderline Category D fluid services. If, for instance, a fluid service is hot enough to be considered hazardous, but is in an area of a facility that sees very little personnel activity then the fluid service could still be considered as a Category D fluid service. One factor I have not included here is the degree of relative importance of a fluid service, or in other words, if a system failed how big of a disruption would it cause in plant operation, and how does that factor into this process. As an example, if a safety shower water system has to be shut down for leak repair the down-time to make the repairs has little impact on plant operations. This system would therefore be of relative low importance and not a factor in this evaluation process. If on the other hand a chilled water system has to be shut down for leak repair to a main header, this could have a significant impact to operations and production. This could translate into lost production and could be considered a high degree of importance. You could also extend this logic a bit further by assigning normal fluid service status to the primary headers of a chilled water system and assigning Category D status to the secondary distribution branches then leak test accordingly. You need to be cautious in considering this. By applying different Category significance to the same piping system it could cause more confusion than it is worth. In other words it may be more value added to simply default to the more conservative Category of Normal. Continuing; if we can consider that there is a high assured integrity value for these piping systems there are two remaining factors to be considered. The first would be: within the above indicated temperature range at what temperature should a fluid be considered hazardous; and secondly, how probable is it that personnel could be in the vicinity of a leak, should one occur. For our purpose here let us determine that any fluid 160F (71C) and above is hazardous upon contact with human skin. If the fluid you are considering is within this temperature range then it has the potential of being 15

considered a normal fluid, as defined in B31.3, pending its location as listed in Table 3-7. Table 3-7 Areas Under Consideration For Cat. D Group Description Yes No 1 Personnel Occupied Space 2 3 4 5 Corridor Frequented by Personnel Sensitive Equipment (MCC, Control Room, etc.) Corridor Infrequently Used by Personnel Maintenance & Operations Personnel Only Access

As an example, if you have a fluid that is operating at 195F (90.6C) it would be considered hazardous in this evaluation. But, if the system is located in a Group 5 area (ref. Table 3-7) it could still qualify as a Category D fluid service. After the above exercise in evaluating a fluid service we can now continue with a few examples of leak test procedures. Using the same symbology indicated in Table 33 these leak test procedures will be categorized as follows: Category T-1: Initial Service Leak Test T-1.1 This Category covers liquid piping systems categorized by ASME B31.3 as Category D Fluid service and will require Initial Service Leak Testing only. 1. If the system is not placed into service or tested immediately after flushing and cleaning, and has set idle for an unspecified period of time it shall require a preliminary pneumatic test at the discretion of the Owner. In doing so, air shall be supplied to the system to a pressure of 10 psig and held there for 15 minutes to ensure that joints and components have not been tampered with, and that the system is still intact. After this preliminary pressure check proceed. 2. After completion of the flushing and cleaning process, connect the system, if not already connected, to its permanent supply source and to all of its terminal points. Open the block valve at the supply line and gradually feed the liquid into the system. 3. Start and stop the fill process to allow proper high point venting to be accomplished. Hold pressure to its minimum until the system is completely filled and vented. 4. Once it is determined that the system has been filled and vented properly, gradually increase pressure

until 50% of operating pressure is reached. Hold that pressure for approximately 2 minutes to allow piping strains to equalize. Continue to supply the system gradually until full operating pressure is achieved. 5. During the process of filling the system, check all joints for leaks. Should leaks be found at any time during this process drain the system, repair leak(s) and begin again with step 1. (Caveat: Should the leak be no more than a drip every minute or two on average at a flange joint, it could require simply checking the torque on the bolts without draining the entire system. If someone forgot to fully tighten the bolts then do so now. If it happens to be a threaded joint you may still need to drain the system, disassemble the joint, clean the threads, add new sealant and reconnect the joint before continuing.) 6. Record test results and fill in all required fields on the leak test form. T-1.2.This Category covers pneumatic piping systems categorized by ASME B31.3 as Category D Fluid service and will require Initial Service Leak Testing. 1. After completion of the blow-down process, the system shall be connected to its permanent supply source, if not already done so, and to all of its terminal points. Open the block-valve at the supply line and gradually feed the gas into the system. 2. Increase the pressure to a point equal to the lesser of one-half the operating pressure or 25 psig. Make a preliminary check of all joints by sound or bubble test. If leaks are found release pressure, repair leak(s) and begin again with step 1. If no leaks are identified continue to step 3. 3. Continue to increase pressure in 25 psi increments, holding that pressure momentarily (approximately 2 minutes) after each increase to allow piping strains to equalize, until the operating pressure is reached. 4. Check for leaks by sound and/or bubble test. If leaks are found release pressure, repair leak(s) and begin again with step 2. If no leaks are found the system is ready for service. 5. Record test results and fill in all required fields on the leak test form. Category T-3.1: Hydrostatic Leak Test T-3.1.This Category covers liquid piping systems categorized by ASME B31.3 as Normal Fluid service.

1. If the system is not placed into service or tested immediately after flushing and cleaning, and has set idle for an unspecified period of time it shall require a preliminary pneumatic test at the discretion of the Owner. In doing so, air shall be supplied to the system to a pressure of 10 psig and held there for 15 minutes to ensure that joints and components have not been tampered with, and that the system is still intact. After this preliminary pressure check proceed. 2. After completion of the flushing and cleaning process, with the flush/test manifold still in place and the temporary potable water supply still connected (reconnect if necessary), open the block valve at the supply line and complete filling the system with potable water. 3. Start and stop the fill process to allow proper high point venting to be accomplished. Hold pressure to its minimum until the system is completely filled and vented. 4. Once it is determined that the system has been filled and vented properly, gradually increase pressure until 50% of the test pressure is reached. Hold that pressure for approximately 2 minutes to allow piping strains to equalize. Continue to supply the system gradually until test pressure is achieved. 5. During the process of filling the system, and increasing pressure to 50% of the test pressure, check all joints for leaks. Should any leaks be found drain system, repair leak(s) and begin again with step 1. 6. Once the test pressure has been achieved, hold it for a minimum of 30 minutes or until all joints have been checked for leaks. This includes valve and equipment seals and packing. 7. If leaks are found evacuate system as required, repair and repeat from step 2. If no leaks are found, evacuate system and replace all items temporarily removed. 8. Record all data and activities on leak test forms. Those three examples should provide an idea as to the kind of guideline that needs to be created in providing direction to the contractor responsible for the work. For leak testing to be successful on your project, careful preparation is key. This preparation starts with gathering information on test pressures, test fluids, and the types of tests that will be required. The most convenient place for this information to reside is the piping line list or piping system list. 16

A piping line list and piping system list achieve the same purpose only to different degrees of detail. On some projects it may be more practical to compile the information by entire service fluid systems. Other projects may require a more detailed approach by listing each to and from line along with the particular data for each line. The line list itself is an excellent control document that might include the following for each line item: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. Line size Fluid Nominal material of construction Pipe Spec Insulation spec P&ID Line sequence number from and to information Pipe code Fluid Service Category Heat Tracing Operating Pressure Design Pressure Operating Temperature Design Temperature Type of Cleaning Test Pressure Test Fluid Type of Test

Manufacturing Practice (cGMP) and FDA requirements. These CFR Titles and FDA requirements drove the need to demonstrate or prove compliance. These requirements can cover everything from verification of examination and inspection, documentation of materials used, software functionality and repeatability to welder qualification, welding machine qualification, etc. The cGMP requirements under 29CFR Titles 210 & 211 are a vague predecessor of what validation has become, and continues to become. From these basic governmental outlines companies, and the pharmaceutical industry as a whole, have increasingly provided improved interpretation of these guidelines to meet many industry imposed, as well as self-imposed requirements. To a lesser extent, industrial projects outside the pharmaceutical, food & drug, and semi-conductor industries, industries not prone to require such in-depth scrutiny, could benefit from adopting some of the essential elements of validation. Elements such as: material verification, leak test records, welder and welding operator qualification records, etc. At face value this exercise would provide an assurance that the fabricating/installing contractor is fulfilling their contractual obligation. The added benefit is that in knowing that this degree of scrutiny will take place the contractor will themselves take extra pain to minimize the possibility of any rejects. And I am not inferring that all contractors are out to get by with as little as they can. Just the opposite is actually true. Most contractors qualified to perform at this level of work are in it to perform well and to meet their obligations. Most will already have their own verification procedure in place. The bottom line is that the Owner is still responsible for the end result. No one wants to head for the litigation table at the end of a project. And the best way to avoid that is for the Owner to be proactive in developing their requirements prior to initiating a project. This allows the spec writers and reviewers the benefit of having time to consider just what those requirements are and how they should be defined without the time pressures imposed when this activity is project driven. Performing this kind of activity while in the heat of a project schedule tends to force quick agreement to specifications and requirements written by parties other than those with the Owners best interest at heart. Validating a piping system to ensure compliance and acceptability is always beneficial and money well spent. Wrapping Up

Developing this type of information on a single form provides everyone involved with the basic information needed for each line. Having access to this line-by-line information in such a concise well organized manner reduces guess-work and errors during testing. Test results, documented on the test data forms, will be maintained under separate cover. Together the line list provides the required information on each line or system and the test data forms provide signed verification of the actual test data of the test circuits that make up each line or system. VALIDATION The process of Validation has been around for longer than the 40 plus years I have been in this business. You may know it by its less formal namesakes walk-down and checkout. Compared to validation, walk-down and checkout procedures are not nearly as complex, stringent, or all inclusive. Validation is actually a subset activity under the umbrella of Commissioning and Qualification (C&Q). It is derived from the need to authenticate and document specifically defined requirements for a project and stems indirectly from, and in response to, the Code of Federal Regulation 29CFR Titles 210 and 211 current Good 17

Before closing out this last of three articles there are just a couple of things I would like to touch on. We had discussed industry Standards earlier and how they are selected and applied on a project. What I didnt cover is the fact that most projects will actually have a need to comply with multiple industry Standards. In a large grass-roots pharmaceutical project you may need to include industry compliance Standards for much of the underground utility piping, ASME B31.1 for boiler external piping (if not included with packaged boilers), ASME B31.3 for chemical and utility piping throughout the facility, and ASME-BPE for any hygienic piping requirements. These and other Standards, thanks in large part to the cooperation of the standards developers and ANSI, work hand-in-hand with one another by referencing each other where necessary. These Standards committees have enough work to do within their defined scope of work without inadvertently duplicating work done by other Standards organizations. An integrated set of American National Standards is the reason that, when used appropriately, these Standards can be used as needed on a project without fear of conflict between those Standards. One thing that should be understood with industry Standards is the fact that they will always be in a state of flux; always changing. And this is a good thing. These are changes that reflect updating to a new understanding, expanded clarification on the various sections that make up a Standard, staying abreast of technology, and simply building the knowledge base of the Standard. As an example, two new Parts are being added to the seven Parts currently existing in ASME-BPE. There will be a Metallic Materials of Construction Part MMOC, and a Certification Part CR. This is all part of the ever-evolving understanding of the needs of the industrial community and improved clarification, through discussion and debate on content. Writing these articles was a form of informational triage for me. There were definite piping topics I wanted to include and others I would have preferred to include, but could leave out without too much of an impact. And then there were the extended discussions on some topics that ultimately had to be sacrificed. This is why some topics were briefer than I would have liked. My attempt at covering such a wide range of discussion on industrial piping was to provide a basic broad understanding of some key points on this topic, not, as I said earlier, to go into great detail on any specific topic.

I hope that in writing these articles I piqued enough interest that some of you will dig deeper into this subject matter to discover and learn some of the more finite points of what we discussed here. I also hope these articles provided enough basic knowledge of piping for you to recognize when there is more to a piping issue than what you are being told. Acknowledgement: My deep appreciation again goes to Earl Lamson, senior Project Manager with Eli Lilly and Company, for taking the time to review each of these three articles. His comments help make this article, and the others, better documents than they otherwise would have been. He obliged me by applying the same skill, intelligence and insight he brings to everything he does. His comments kept me concise and on target. About the author: W. M. (Bill) Huitt has been involved in industrial piping design, engineering and construction since 1965. Positions have included design engineer, piping design instructor, project engineer, project supervisor, piping department supervisor, engineering manager and president of W. M. Huitt Co. a piping consulting firm founded in 1987. His experience covers both the engineering and construction fields and crosses industrial lines to include petroleum refining, chemical, petrochemical, pharmaceutical, pulp & paper, nuclear power, biofuel, and coal gasification. He has written numerous specifications, guidelines, papers, and magazine articles on the topic of pipe design and engineering. Bill is a member of ISPE (International Society of Pharmaceutical Engineers), CSI (Construction Specifications Institute) and ASME (American Society of Mechanical Engineers). He is a member of three ASME-BPE subcommittees, several Task Groups, an API Task Group, and sets on two corporate specification review boards. He can be reached at: W. M. Huitt Co. P O Box 31154 St. Louis, MO 63131-0154 (314)966-8919 wmhuitt@aol.com www.wmhuitt.com

18