Conflicts and solutions for complying with grounding revisions of the NEC for cathodically protected facilities.
During the 2005 National Electrical Code (NEC) review cycle, a relatively simple revision was made that created a significant change on the cathodic protection (CP) systems of petroleum and chemical facilities (plants). The 2005 NEC made it clear that there must be an interconnection of all concrete-encased steel reinforcing bar (rebar) to the facility grounding electrode system.
For facilities using copper grounding electrode systems (and also having cathodic protection as part of their corrosion mitigation and integrity programs), this revision may require a facility to change their entire cathodic protection design and operating philosophy. It will definitely require greater coordination and cooperation between the various engineering and construction disciplines.
2005 Code changes
In NEC Sec. 250.50, it is required that all grounding electrodes present be bonded together to form the grounding electrode system. The types of grounding electrodes that must be bonded together, if present, are listed in 250.52(A)(1) through (A)(6). They include: 1) metal underground water pipe, 2) metal frames of buildings or structures, 3) concrete-encased electrodes (usually rebar), 4) ground rings, 5) rod or pipe electrodes, and 6) plate electrodes. This article will focus on concrete-encased electrodes, copper ground rings, and the potential problems this may create by bonding without consideration of potential corrosion.
Prior to the 2005 edition of the NEC, the first paragraph of Sec. 250.50 started with the phrase: “If available on the premises…” all of the types of grounding electrodes listed in the preceding paragraph were required to be bonded together. More times than not, industrially oriented engineers and designers took the phrase “if available” to mean that since the concrete structures were usually poured before the electricians were onsite, then the rebar wouldn't be available. Therefore, they assumed it wasn't necessary to bond the rebar to the other grounding electrodes and make the rebar part of the grounding electrode system. In other words, the way the NEC used to be worded seemed to give the designer some latitude as to whether to bond all of the potential grounding electrode types together.
Right or wrong, this was — and still is — the typical attitude when designing plant grounding electrode systems. If you're a design engineer who tries to follow the NEC, however, removal of the words, “If available on the premises…” has a rather profound meaning. Now it becomes clear that one must bond the copper ground ring to the concrete-encased rebar (and all other electrodes), unless the rebar is in an existing structure. With a literal interpretation, all concrete structures within a process unit, with the equivalent of more than 20 feet of ½-inch (or larger) rebar in it, must be bonded to the copper ground ring and all other electrodes. This includes pump foundations, pile caps, spread footings, concrete piers, and tank bottoms. For new process units, this adds considerable initial costs and probably even more so later, due to the resulting corrosion currents.
In today's refineries and petrochemical facilities, a variety of philosophies are used. Many plants want a large copper grid or grounding rings in each process unit. These “ground grids,” “ground rings,” or “ground loops,” as they are often referred to, are typically installed with 10-feet or 20-feet copper-clad rods spaced at an average distance of 50 feet to 100 feet with a copper wire run between the rods. The copper wire between the rods is usually bare, although occasionally insulated cable is used. The conductor sizes are normally at least 4/0 AWG and can be as large as 500 kcmil. Equal or smaller size taps from the ground ring are usually bonded to structural steel aboveground throughout the unit and to all motors, transformers, and other equipment such as tanks and vessels. The intent is to create an equal potential plane while also creating a low resistance to earth for ground faults and lightning strokes. For the purpose of this article, this ring, grid, or loop will be referred to as a “ground ring.”
While most plant installations in the United States today have a copper ground ring, there are some facilities that prefer only an equipment-grounding conductor run with the branch circuits and feeders to electrical equipment (either in the form of metal conduit, cable tray or wire, or a combination thereof), thinking that the circuit equipment grounding conductors combined with structural steel and piping being in contact with each other is all that is required to keep everything at an equal potential. For system grounding in these cases (at a transformer secondary, for instance), there may be one or two ground rods installed to supplement the grounding electrode system.
Then there are other facilities that prefer something in between these two methods, like small, partial ground rings around specific equipment, such as transformers and substations, bonded together, and equipment-grounding conductors run with branch circuits and feeders. This discussion will focus only on those systems with large copper grounding rings, because this is where the probability for buried steel corrosion is the greatest.
Bonding and grounding materials
When bonding the copper ground ring to the concrete-encased rebar, several approaches can be used; however, an effective path must be provided. One may choose to use exothermic welding on the rebar (Photo) or listed compression connectors — and attach a copper wire “pigtail” to the rebar. (It's important to note that all grounding must be done with either exothermic welding or listed materials per the NEC.) This wire can then be run just outside the foundation forms and extended at a later time to the ground ring (Fig. 1).
Another method is to install a copper ground pad flush on top of the concrete slab and bond it to the rebar prior to the pour. During construction of the ground ring, taps can be installed and bolted to the copper ground pad (Fig. 2).
An additional technique is to leave a piece of rebar protruding from the concrete into the soil to tap onto during ground ring construction (Fig. 3). Depending on the type of soil and moisture content, however, the exposed rebar may not be “permanent,” because it may rust and deteriorate fairly quickly.
Yet another means of bonding is to weld the anchor bolts (where used) to the steel rebar before the concrete is poured. Then, during construction of the ground ring, a tap can be run up to a steel tab welded to the structural steel at a convenient location above ground (Fig. 4).
Regardless of which method is used, they all require either mobilizing the electrical contractor long before it would normally be required or adding work on the civil contractor's part. The additional costs for supplying a means to bond the copper ground ring to the rebar are considered negligible when compared to the corrosion that will occur in the rebar when the connection is made without means to prevent corrosion of the buried steel.
Grounding, corrosion, and the IEEE “Green Book.”
In IEEE Std 142 “Green Book,” Sec. 4.2.3 discusses the use of rebar as a grounding electrode. Corrosion is discussed in Sec. 4.4.5. This section states: “The basic objectives of a sound electrical grounding system are safety of personnel, reliability of equipment operation, fault current return and to limit transient over voltages. After these objectives have been satisfied, the effect of the grounding installation on corrosion must be considered. Systems, equipment, and lighting sometimes unknowingly contribute to the corrosion of underground conductors, structures, and piping.” In the same chapter of the “Green Book,” it also explains how corrosion works when copper and steel are connected.
Unfortunately, the NEC doesn't take matters such as corrosion into account. Making the steel rebar part of the grounding electrode system is a very good idea for safety — at least in the early life of the steel rebar. After a few years, if corrosion isn't addressed, corrosion of the rebar may cause cracking and spalling of the concrete structure such that its integrity may be compromised. The integrity of the rebar grounding “safety” electrode at this point becomes questionable. Thus, in cases where large amounts of copper will be installed below ground and bonded to the buried or concrete-encased steel, a CP system should be installed — or a different material than copper should be used for the ground ring.
Cathodic protection issues
The National Association of Corrosion Engineers International (NACE) is an organization similar to the IEEE, except it deals primarily with corrosion issues. According to NACE, “Cathodic protection is achieved by making the structure the cathode of a direct current circuit.” In other words, cathodic protection is the practice of ensuring that all current flow between a protected metallic structure and its electrolytic environment is from the electrolyte to the structure. No current flow from the structure means no metal loss from the structure.
The major root problem of the recent NEC grounding revision arises from the modern construction practices of using copper for grounding purposes and steel for structural purposes. Steel and copper have very different energy levels with steel being the more negative and anodic of the two. This has probably contributed to the practice of using copper for grounding as the copper components appear to last forever, when in actuality, the copper has been cathodically protected by the steel anode to which it has been connected. Bonding the two together creates a corrosion battery with a driving voltage of approximately 0.5VDC. Only the circuit resistance is variable and controls the amount of current/resulting metal loss. If the resistance to earth of “concrete-encased electrodes” is sufficiently low to make bonding to them an improvement to the facility grounding, the reader can be assured that ionic current will flow from the steel anode to the copper cathode, and the rebar will deteriorate (Fig. 5).
In addition to the dissimilar metals problem, there are other design complications from the grounding revision. For example, the surface area of metal to be cathodically protected will be substantially increased. Cathodic protection designers base their preliminary design calculations on needing a certain number of milliamps of CP current per unit area of bare metal. While rebar is concrete encased, grounding components are bare by design. Even if foundation rebar and/or facility grounding were included in the CP scheme beforehand, which is not uncommon, the new combination will increase CP current requirements by several times. Another problem is that of “stray currents.” CP designers constantly fight the battle of distributing their intended CP currents to the intended structures. This is an unappreciated task. Metallic piping systems such as firewater, potable water, demin water, storm water, and “oily water drain” systems may be routinely considered. Often falling through the consideration cracks are plant air and other utility gases, electrical trays and ducts, instrument and communication cables, nearby utility grounding, and even fencing. Bonding rebar to ground rings (or other grounding electrodes) can inadvertently lead to numerous unintended stray current paths.
CP and rebar/grounding bonding solutions
The first and simplest practical solution to the new rebar grounding requirements is to determine if the facility can be designed and/or operated without external CP. This is a reasonable exercise and can have favorable results if piping can be routed above grade or can be constructed using nonmetallic materials. (The use of corrosion-resistant alloys for piping is cost prohibitive and perfect, never-deteriorating coatings simply do not exist.) Aboveground storage tanks (ASTs) will have to be constructed on solid concrete slabs or with dielectric liners and concrete will have to be chloride free or nonmetallic reinforcement will have to be used. Although it rarely happens, it should be mentioned that there is the rare case where CP can be eliminated because the facility is being built in a location with extremely high soil resistivity. Even then, eliminating CP generally requires that no lower resistivity fill be brought in, that copper grounding equipment be replaced with stainless steel, and that possible future soil contamination, such as the use of de-icing salts in the winter, be eliminated. It should also be mentioned that sometimes a short design life — 10 years or less — will justify the elimination of CP and protective coatings and will rely instead on corrosion allowances and the substitution of stainless steel for copper grounding to achieve the integrity goal.
The second solution activity is to determine if the facility can be cathodically protected as a whole. With few exceptions, this is generally the case. Again, ASTs will still have to be constructed on solid slabs or with dielectric liners; however, steel reinforcement of concrete foundations can be cathodically protected. Sometimes, unusual facility instrumentation requirements or locations with extremely high soil resistance require limited area CP systems. When the facility is suitable for total CP, several remote or deep anode beds can generally be installed to provide suitable total current and current distribution. Sometimes, supplemental limited local distributed anodes will have to be installed in congested areas or in areas of corrosion “hot spots.”
When the above “strategic” solutions are not possible, targeted solutions must be employed. These will involve a combination of the use of non-copper grounding materials, specialized isolation equipment, separate grounding systems, nonmetallic materials for a variety of purposes, combinations of local and area CP systems, and possibly the use of automatically controlled rectifiers. This type of facility treatment will require complex structure-to-soil potential and/or voltage gradient surveys to determine the existence or mitigation of “stray current” situations. Personnel safety and mechanical integrity can both be accomplished, but the methods may be non-traditional — and engineering and procedures may need to be site specific.
Providing CP to “all” of a facility will initially appear to be the most expensive of the strategic solutions. However, when the “lifetime” monitoring and maintenance costs of targeted facility CP systems are considered, applying CP to “only what we need or have to have to meet regulatory requirements” is often the most expensive course of action.
Trimble is a NACE International Certified Cathodic Protection Specialist, an instructor for the NACE CP1 and CP2 courses, and a member of numerous NACE technical committees. He is also a member of the ASME B31.4 Committee “Pipeline Transportation Systems for Transportation Systems for Liquid Hydrocarbons and Other Liquids.” He can be reached at Whitt.Trimble@fluor.com. Guidry is a senior electrical design supervisor at Fluor, Sugar Land, Texas. He is a licensed master electrician in Texas and a certified ICC/IAEI electrical inspector. An active member of Code Panel 11 since 1999, he is also a member of the International Association of Electrical Inspectors and the NFPA. He can be reached at Eddie.Guidry@Fluor.com.
Based on “The Conflicts and Solutions to Complying with the Grounding Revisions of the 2005 National Electrical Code for Cathodically Protected Facilities” by Whitt L. Trimble and Eddie Guidry, which appeared in the IEEE 2007 PCIC Record of Conference Papers. © 2007 IEEE.