Proceedings of the 2012 9th International Pipeline Conference IPC2012 September 24-28, 2012, Calgary, Alberta, Canada

IPC2012-90193

DEVELOPING GUIDELINES FOR ON-STREAM MECHANICAL CLEANING OF A COMPLEX PIPELINE NETWORK

ABSTRACT On-stream pipeline cleaning is essential to pipeline operations in terms of maintaining both pipeline efficiency, as well as integrity. Running effective brush cleaning pigs through pipelines will remove accumulated solid debris as well as smoothen the pipe wall. This will reduce pipeline pressure as well as increase product flow. A pig with sealing discs will push out accumulated condensates in gas pipelines and will prevent internal corrosion. The concept is simple, but the implementation is challenging for a complex pipeline network such as the one covered in this paper. The network consists of different types of services both upstream and downstream including oil, gas, and refined products. The pipeline sizes range from 3 to 56. The age of pipes vary from 50 years old to new installations and accordingly, their conditions and cleanliness vary. Although pigging is largely based on experience and it is more of an art than a defined science, the complexity of the network needed a systematic approach to develop meaningful and useful guidelines. The approach included conducting gap analysis of current pigging practices in the company, interviews with the senior Operators and subject matter experts, and reviewing the latest literature including manufacturer brochures. The developed guidelines include selection criteria, design, material, as well as measurement effectiveness. The objective of the paper is to present the approach taken to develop the guidelines, as well as to give an overview of the guidelines themselves. KEYWORDS On-stream cleaning, pigging, corrosion.

consists of more than 20,000 km piping carrying different types of services, both upstream and downstream. These include oil, gas, and refined products. The pipeline sizes range from 3 to 56. The age of pipes varies from 50 years old to new installations and accordingly, their conditions and cleanliness vary. This network faced several safety and operational-related challenges, as well as a high internal corrosion rate. This led to the requirement to develop a more effective Pipeline Cleaning Program. It was decided to explore practical means to develop a cleaning program on the basis of industry best practices. It has been reported that a good pigging program results in considerable reductions in corrosion rates [1] There are many references that give pigging guidelines [26]. However due to the fact that pigging is largely based on experience and it is more of an art than a defined science, the available guidelines are quite general. Moreover, the complexity of the pipeline network needed a systematic approach to develop meaningful and useful guidelines. The approach included conducting gap analysis of current pigging practices in the company, interviews with the senior Operators and subject matter experts, and reviewing the latest literature including manufacturer brochures. The objective of the paper is present the approach in developing the guidelines as well as giving overview of the guidelines themselves. The developed guidelines include selection criteria, design, material, as well as measurement effectiveness. Its not a common practice in the industry to standardize pipeline cleaning practices as this work is often considered to be proprietary and exclusive to service providers. The main goal of the guidelines is to develop a program that achieves a satisfactory level of cleaning quality.

1.0 INTRODUCTION The Pipeline network transports hydrocarbons to export terminals, processing plants and domestic users. The network

Copyright 2012 by ASME

3.0 METHODOLOGY TO DEVELOP THE GUIDELINES The desired pigging guidelines should cover a wide range of various aspects. It must be good for general cleaning purposes as well as specific pigging application such as removal of black powder from sales gas pipelines. It should cover the range of pipe sizes as small pipe sizes 3" to as large as 56". The different types of pigs including foam, spheres, mandrels and solid cast must be addresses. The Company has already been doing pigging operations for more than 50 years, but it was based on bits and pieces of guidelines, people experience, and sometimes trials. Therefore, it was necessary to adopt a methodology that follows a systematic approach to achieve a-best-in-class pigging guidelines. The methodology consists of the following steps: conducting gap analysis of the current pigging practices in the Company, conducting literature review of best practices of pigging including manufacturer recommendations [2-6], interviewing subject matter experts (SMEs) in the Company, developing a draft guidelines, reviewing the draft with the Company SMEs and the pigging manufacturers. The methodology showed success through the positive feedback received from the reviewers. The reviewers indicated that the guidelines are comprehensive and would certainly improve the pigging operations in the Company. They provided very valuable suggestions in the details of selection, design and material which were incorporated in the final version of the guidelines. 3.0 GAP ANALYSIS This gap analysis is the basis to initiate a task to develop Pigging Guidelines for On-Stream Cleaning of Pipelines document to set minimum requirements for the selection, design, fabrication and materials of pigs as well as to maintain effective cleaning program that is results oriented, dynamic yet achievable and more importantly measurable addressing basic needs of such critical activity. 3.1 Current State There is a lack of standardized guidelines to managing onstream pigging activities, both in the industry, and within the Company. Therefore, cleaning initiatives varied due to manpower availability and based on the urgency of performing cleaning runs. This urgency is most often driven by in-lineinspection (ILI) requirements. In some cases, pigging runs are scheduled to mitigate reported corrosion related upsets such as increased water settlement. Pig design and fabrication practices have varied and for the most part have not followed industry best practices due to limited resources. Pigs, fabricated in an in-house facility, have been designed and fabricated with limited capabilities to do so. Overall program management and performance measurement is another area targeted for improvement. Currently, performance is measured based on the total number of pigging runs conducted annually compared to the number of scheduled runs. The performance measures are not detailed

down to the individual pipeline segment level. At times, this can be a misleading indicator as the total number of cleaning runs might meet the target. However, some pipeline segments get additional or unplanned runs due to ILI inspection requirements or recorded upsets. At the same time, other pipelines are not adequately pigged. The pig run scheduling data is not currently available in a shared online platform and therefor the data is difficult to revise or update when required. Based on the discussion above, it has been agreed that current state of cleaning program needs to be enhanced to achieve a minimum requirement of such important pipelines activity. This desired state is described below against the respective current state categorized as follows; 1) Gaps in the pig design and fabrication 2) Gaps in the pigging program 3) Program reporting and performance measurements 3.1.1 Gaps in the pig design and fabrication Several gaps were identified in the pigs that are fabricated internally in the Company. The main gaps relate to the guide disc, cleaning brushes and elastomer materials. Guide discs Most of the cleaning pigs fabricated in-house, experience a diving, or nose down phenomena due to the lack of good support at the front of the pig. This is often because of inappropriate polyurethane material being used for the guide discs. The current practice is to cut both the guide discs, as well as the sealing discs, from PU sheets that have low hardness values (705 Shore A) and low thickness (0.5 to 1). While this hardness and thickness are good for sealing discs, they are not sufficient for guide discs. Stable support is required for effective scraping which can be only maintained by the right selection of supporting guide discs. Providing enough guide discs for scrapers will help the sealing disks to work efficiently and provide drive to the tool. A scraper that is far off from the centerline, due to inadequate guide disks, is likely to cause inadequate seal, disc damage or worse, a stuck cleaning tool (Fig. 1).

Figure 1. Scraper Stuck due to Seal Lipping off

Copyright 2012 by ASME

Brushes Most of the internally fabricated pigs have an old brush design where the bristles are too brittle and fall off in long pipeline segments. The newly ordered pigs come with brushes (when required) but selection of brush type is often driven by availability and not by specifications. Recently, speciallydesigned brushes were piloted successfully, but have not been rolled out for general use. The enforcement of a governing best practice will allow for better brush quality and performance. Elastomer Materials Pig cup and disc material is typically manufactured from polyurethane (PU) material with adequate abrasion and tear resistance. Such properties tend to fail with however with significant temperature increase. Neoprene, nitrile, EPDM, and Viton however are preferred over standard PU for high temperature pipelines. Pig discs and cups made from polyurethane behave satisfactorily for the vast majority of the applications. Due to lack of standardized material specification, procured PU materials are governed by availability in local markets. This had allowed for the utilization of sub standardized PU sheets. PU materials are bought in bulk quantities, and are often used for multiple runs. This requires environmentally controlled storage as PU materials are vulnerable to humidity, temperature and exposure to direct sun light. The in-house facility does not provide sufficient controlled storage which further limited the performance of the PU material. 4.1.2 Gaps in the pigging program The main gaps identified in the pigging program are determining the cleaning frequency and the pigging runs records. Cleaning Frequency The most important aspect of effective cleaning programs is cleaning frequency. There is inconsistency in determining the cleaning frequency as it is only relying on the experience of the responsible engineer. The annual cleaning plan for each piggable section of the network is mainly defined based on the type of fluid running in the line. In some instants, some pipeline sections receive special attention (i.e., through increasing cleaning frequencies) driven by the ILI inspection date. On the other hand, standardized approach to decide upon the cleaning frequency and when it shall be changed does not exist. Some pipelines are being cleaned on a constant frequency without assessing the need to increase the frequency due to changing operational parameters (such as water ingress) or reported corrosion growth. Guidelines for cleaning frequency were partially developed for some crude oil pipelines. However, this was not generalized to cover wider range of the hydrocarbon services such as sales gas that experiences black powder.

In brief, the current cleaning frequency has the following shortcomings: 1) Fixed frequency for each pipeline decided by service. 2) Plans are not dynamic enough to address pipeline upset and cleanness issues to minimize the associated impact on pipelines integrity on timely basis. Pigging runs records The pig cleaning plans are set one time at the beginning of the year. However, there is no controlled procedure for updating the plan. In many instances, the changes in the plan during the year are not updated. Even in case of updates taken place, the updated plan is not easily retrievable as it is not available in a centralized database. Runs performance and cleaning jobs quantifiable measures do not exist. Reporting the results of pigging runs is not practiced. The current Key Performance Index (KPI) which measures the percentage of total pigging runs to the planned run could be misleading. While some pipeline segments get more than planned pigging runs due to ILI inspection period or recorded upsets, others are not adequately pigged. 4.2 Impact of the in-effective pigging operations An internal corrosion assessment study revealed a dramatic increase in the internally corroded pipelines for a studied group of pipelines representing a major part of the network (Fig. 2). Comparison between two consecutive ILI runs showed an increase of 60% in the total number of internally corroded joints. The majority of the indications were the result of new internal corrosion anomalies. This supports the argument that current pigging program is ineffective. Moreover, It has been repeatedly observed that corrosion pits may fill with metallic debris. This makes the pits harder to be interpreted and sized correctly by ILI tools and can give misleading results. To avoid this, pigs should be run prior to each ILI run and these pigs should be equipped with effective brushes to remove debris from the corrosion pits. If the pipeline is not properly cleaned before the ILI run, the whole inspection run might not be of acceptable quality and may have to be repeated. In addition to adding operational disruption, ILI reruns can be costly in terms of both direct and indirect costs.

Figure 2. Comparison of corrosion findings between 2 ILI runs

Copyright 2012 by ASME

4.2 Desired State The desired state can be summarized in two simple goals: 1. 2. Having scarpers that fit for purpose in terms of selection, design, and material (Fig. 3). Having effective and optimized pigging program covering pigging plan, frequency, and KPI measures.

a)

Maintaining the pipeline integrity and minimizing internal corrosion by removing solid deposits and stagnant water.

b) Improving the pipeline efficiency by smoothening the pipe and reducing pressure drop. c) Preparing the pipe for in-line inspection (intelligent pigging) by removing debris or foreign matter of the pipeline.

To bridge the gap between the current state and desired state, the following were recommended as action plan: 1) Establishing Guidelines that act as a minimum requirement for scraping design, fabrication, materials, frequency of cleaning and program measurements. (discussed with some details in section 4.0 of this paper) 2) Developing Online Platform for the pigging plans with dynamic viewing and editing features. Also, the ITenabled platform shall contain performance measurements monitoring capabilities (on-going IT project, not in the scope of this paper).

d) Protecting downstream facilities by reducing the received amount of water, salt, deposits, etc. Selection of the pig depends mainly on the purpose of pigging. Table 2 gives a summary of the recommended scraper type depending on the scraping purpose. The selection is not a oneto-one relation. Different types of scrapers can be used for the same purpose. On the other hand, one type of scraper can serve several purposes. The table shows that mandrel type scraper equipped with the right attachment can be used for several purposes.

4.2 Pigging frequency The frequency of pigging depends on the specific pipeline parameters and scraper run history. The frequency should increase with low fluid velocity, high amount of deposits, large amount of water, and high corrosion rate. A number of scraping runs and an accumulative database is necessary before the optimum frequency is achieved. Therefore, a fixed-frequency table was developed to be used for new pipelines or when there is a lack of data of previous pigging runs. Table 3 shows sample of the fixed-frequency table. 4.3 Design The guidelines focused on the mandrel type pig as it is the most common and most versatile and due to the relative simplicity of the other types of pigs. The design requirements of mutli-bolt steel body mandrels (Fig. 4) are discussed in details in the guidelines. The design requirements cover the body as well as the attachments: guides, sealing elements, support elements, and cleaning elements. For example, the guide outside diameter must equal to 99% of the pipeline inside diameter, while the sealing disc outside diameter should be between 103% and 108% of the pipeline inside diameter. A table of typical dimensions is provided in the guidelines; an example of which is given in Table 4. The issue of dual diameter pipelines was also discussed and several options are given such as conical cups, slotted discs, or proprietary designs. 4.4 Material The scraper body and all of its components shall be suitable for the pipeline content. The scraper material used shall not absorb or contaminate the pipeline content. Although this seems trivial, it could be overlooked cases like sour service or high temperature. However, the most important part covered in

Figure 3. Desired state: pig fit for purpose.

4.0 OVERVIEW OF THE GUIDELINES The objective of this section is to give an overview of the guidelines and some examples of its details. The guidelines contents include: scope, selection of pig, frequency of pigging, design, material, maintenance and storage, effectives measures of a pigging program. 4.1 Selection of pig The different types of pigs are shown in Table 1 in the Appendix. Use of the correct type of scraper and a proper scraping program will help in:

Copyright 2012 by ASME

this section is the mechanical properties of the elastomers used for the guide and sealing discs.

4.6 Effectiveness measures The guidelines established clear definitions for measuring the effectivenss of a single pigging run as well as pigging program. A pigging run is considered succesful if: The objective of the pigging run has been met (liquid removed, debris cleaned, increase of flow rate, reduction in pump or compressor power, improved product quality, etc.). The pig is received in good conditions.

Bolts

Spacers Sealing discs Fixed flange

Support (guide) disc

Sliding end flange or plate

Mandrel body, metal cylinder

A cumulative database of pigging run post-reports must be available to measure effectives of the scraping program. The effectivenss of pigging program may be measured by: The amount of liquid or debris received at each scraper run. The cleaniness of the scraper after each run. The reduction of corrosion rate measured by corrosion monitoring devices or successive runs of in-line inspection. The Compliance percentage of the actual performed scraper runs compared to the planned scraper runs. The Optimized scraping cost by optimizing scraper frequency per pipeline specifics.

Figure 4. Illustration of flanged mandrel assembly

The guide discs should be made from polyurethane material with 75-85 shore A hardness while the sealing disc should be made from polyurethane material with 65-70 shore A hardness. The elastomer material shall have documentation proving they meet the mechanical properties given in the guidelines. 4.5 Maintenance and storage The guidelines give procedures for the maintenance and storage of pig and components. To help in maintaining records of the pig and its conditions, each pig shall have a unique identification number and shall be logged in the pig inventory sheet. The pig body and components should be cleaned after each pigging run. The pig body and components should be carefully inspected after cleaning for damage, particularly the elastomer parts. The results of inspection should be documented in the pig maintenance history. For storage of the pigs, the components parts should be protected against atmospheric corrosion. Elastomer components should be stored in shaded areas away from direct sun light. Storage time of polyurethane components should be minimized by planning inventory to avoid large stock. Reordering on a regular basis and a strict system of first-in and first-out should eliminate potential problems due to material failures.

5.0 CONCLUSION Comprehensive pigging guidelines were developed for a large pipeline network through a systematic approach that was described in this paper. Developing the guidelines is a very important milestone in having a successful and effective pigging program, but is not the ultimate goal. The next step is implementation and continuous improvement. As was the case with the development, the implementation will follow a systematic approach for improvement using six sigma method (DMAIC). DMAIC is short for define, measure, analyze, improve and control. A group of pipelines will be selected for implementation of the guidelines. Measurement of the current corrosion rate will be conducted before and after implementation of the pigging guidelines. This will be followed by Analysis of the data and improvement of the guidelines. Finally, control measures of the pigging program will be enforced. ACKNOWLEDGMENTS The authors would like also to acknowledge Saudi Aramco for their support to present this paper. We would like also to acknowledge the valuable contribution and feedback of experts who reviewed the guidelines in particular Jamal Al-Khamis, and Gordon Rentz.

Copyright 2012 by ASME

REFERENCES 1. Fernandes Jr., A. A., Dantas, F. R., and A. de Souza Lima, 2006, The Results of Reduction in Corrosion Rates Using Cleaning Pigs, Proceedings of IPC 2006, Calgary, Sep 2529, 2006. 2. Jim Cordell and Hershel Vanzant, 2003, The Pipeline Pigging Handbook, 3rd Edition. 3. Scraping Products & Services Association, 1995, An Introduction to Pipeline Pigging. 4. John Tiratsoo, Editor, 2003, Pipeline Pigging & Integrity Technology, 3rd Edition. 5. H. Lindner, 2006, A New Cleaning Approach for Black Powder Removal, Presented at the 2006 seminar of Piping Products and Services Association, Aberdeen. 6. P. Fretwell, 2007, Developments in Mechanical Production Cleaning of Pipelines, Presented at the 2007 seminar of Piping Products and Services Association, Aberdeen.

Copyright 2012 by ASME

Appendix A: Tables extracted from the pigging guidelines Table 1. Types of pigs Type Picture Description Foam scrapers are made of open cell polyurethane foam material of light, medium, or heavy densities. Foam scrapers may be a simple cylindrical shape with a flat front and rear or a bullet shape. The foam scraper can be bare, or coated with roughed surface for better cleaning. The foam scraper has advantages of being compressible, expandable, light weight and flexible which allows passage through abrupt changes in piping as well as reduced openings such as valves. However, they have the disadvantage of being used for one-time and for short length of runs. Furthermore, Foam scrapers may degrade into pieces or take unexpected flow directions in tees. Sphere scrapers can be manufactured from foam or solid cast to the spherical shape. Moreover, they can be manufactured from elastomer and inflated with liquid. They are almost exclusively used for sealing. They are widely used for removing condensates and batching operations. Mandrel scrapers have a metal body with a number of component parts that are mounted on it. As a minimum, the mandrel scraper shall have support elements to centralize the scraper and sealing elements to provide the differential pressure to propel the scraper. Additional elements may be attached for improvement in performance such as brushes, blades, magnets, etc. This gives the mandrel scraper its main advantage as the component parts mounted on the body may be replaced or reconfigured as the need arises.

Foam

Sphere

Mandrel

Solid Cast

Solid cast scrapers are usually made of polyurethane, but can be manufactured from other rubber elastomers. They are mainly used as sealing scrapers although wraparound brushes can be added for cleaning function. The solid cast scrapers are normally limited to sizes up to 16.

Copyright 2012 by ASME

Table 2. Selection of scraper type depending on the pigging purpose foam Type plastic disc plastic blades metal blades foam brush Purpose Wax removal soft dirt removal hard deposit removal Ferrous material removal liquid removal metal brush foam plain Sphere mandrel Magnets Solid cast Soild cast plain Soild cast brush

Table 3. Pigging fixed-frequency depending on service Service Sweet crude Corrosion reason Objective of scraping Presence of water due to Remove water low velocity (liquid) Presence of salt (catalyst Remove salt (fine to corrosion) particles) Bacterial growth Clean pit (corrosion channel at 6 clock position) Disturb biofilm (bacterial growth) Distribute corrosion inhibitor Minimum expected General cleaning internal corrosion and do not have corrosion inhibitor due to: Minimum water content High flow rate and velocity Frequency 3 months, flow <3 ft/s, 6 months, flow >3 ft/s

Refined product and NGL

1 year

Copyright 2012 by ASME

Table 4 Typical dimensions of Bi-Di pig L

A Size Type OD1

B Thickness 2 3/8" 1 1/2" 1 1/2" 2 3/5 2 13/20 2 13/20

C Center Hole 32 1/4" 32 1/4" 32 1/4" 35 35 35

D Bolt Circle 36.0" 36.0" 36.0" 39.0 39.0 39.0

L Length Bolt #-size of Scraper 24 - 5/8" 24 - 5/8" 24 - 5/8" 28- 3/4 28- 3/4 28- 3/4 78" 72"

Guide Disc 47 1/4" 48" Sealing Disc 48.0" Spacer Disc 38 1/2" Guide Disc 51 1/4 52 Sealing Disc 52 Spacer Disc 42 1/2

Copyright 2012 by ASME