Electrical safety and wind turbine technology
It was the fall of 1973. I‚Äôd been waiting in line to get gas for an hour, when a pimply teenaged pump attendant promptly dragged out a freestanding sign proclaiming, ‚ÄúNo More Gas ‚ÄėTil Tomorrow!‚ÄĚ Ah, the good old days. For the first time, most of us old enough to remember those days were learning that America was not a self-sufficient country when it came to energy. OPEC had closed its oil spigot to America. We were mad as hell, and we needed to do something about it.
In reaction to this frightening reality, government and businesses scrambled for solutions. One approach was to mainstream the idea of ‚Äúrenewable energy.‚ÄĚ Wind energy seemed like a great answer to energy independence. Unfortunately, hastily conceived government construction subsidy programs lead to a flurry of unprofitable windmill installations. The poorly realized scale of turbine technology (i.e., too small), its bulky untested designs, and ‚ÄĒ most importantly ‚ÄĒ the normalization of oil prices in 1974 saw the wind energy industry quickly slide out of the spotlight of national energy policy. As a result, wind research and development in the United States all but ceased.
Before and after
Like most of the general public, back in those days, the term ‚Äúwindmill‚ÄĚ conjured an image in my mind of tulips and quaint Dutch maidens dancing across a background landscape dotted with rustic windmills. As an apprentice electrician, I saw those stumbling first attempts as pretty marginal ‚ÄĒ not to mention inconsequential ‚ÄĒ to my nascent career.
While wind technology languished on the back burner in this country, countries like Denmark, Holland, Spain, Germany, and Japan stuck with it. Their persistent research and development put them in the forefront of today‚Äôs wind power industry. However, America has done a pretty good job over the last few years of playing catch-up.
Today, we‚Äôre seeing a massive effort to bring wind energy back to the forefront of power production in this country. Unlike the not-ready-for-prime-time attempts of the ‚Äô70s, wind power is being developed with a technology that has a proven track record. In fact, the United States currently has more functioning wind turbines installed than any other country. This title will undoubtedly be short-lived, however, as China is rapidly catching up ‚ÄĒ and is expected to pass the United States in the next few years. Not surprisingly, China is already the largest manufacturer and exporter of wind turbines in the world.
In 2009, a Harvard study, ‚ÄúGlobal Potential for Wind-Generated Electricity,‚ÄĚ made the remarkable claim that there are enough developable wind energy locations in the lower 48 states to provide more than 16 times the total energy requirements of the entire United States. This study‚Äôs results are based on current available technology (i.e., 2MW to 3MW turbines). That‚Äôs right, 2MW to 3MW turbine/generators are current technology.
Here‚Äôs where I want to explain the title of this article. The scale of current wind turbine technology is hard for most people to fathom. Today‚Äôs wind turbines are a far cry from the stereotypical Dutch windmills most of us harbor. Instead, they‚Äôre massive industrial marvels. For example, if you‚Äôve ever been on the tarmac of a major airport and found yourself next to a Boeing 747 jumbo jet, you might have wondered how the heck that mountain of metal gets off the ground. It takes wings that are a colossal 211 ft from tip to tip. In comparison, the wing span (diameter of the circle drawn by turning turbine blades) of a typical 2MW wind turbine is 262 ft to 295 ft (Photo 1). The fundamental economic viability of wind farms depends on this enormous scale.
Operating voltages and system configuration
Turbines generate at various voltages, depending on the manufacturer and model. These generator output voltages can range from 480VAC to 1,000VAC. Several manufacturers generate at 690VAC. These turbines are especially difficult to inspect because they defy easy categorization in the NEC or IEEE standards. The NEC does not list 690V as a ‚Äúnominal voltage.‚ÄĚ Rules for more than 600V are distinguished from rules for 600V and under. IEEE standards for power circuit breakers define low voltage as up to 635V.
Turbine voltages are stepped up at each tower to an industry standard of 34.5kVAC. This 34.5kVAC is then daisy-chained by underground cable runs called collector circuits. Most wind farms comprise multiple strings of turbines. Strings can be made up of as few as four turbines to as many as 16. Typically, three to six strings will terminate at a main bus in the substation powerhouse. Within the substation yard, voltage will be stepped up to the local utility‚Äôs transmission voltage (i.e., greater than 69kV). Although 60MW to 80MW substations are most common, farm sizes as large as 500MW are in the pipeline.
One of the principal dangers on wind farms relates to the extremely high fault currents that are available when ground faults and short circuits occur on the system. One main source of fault currents is the numerous generators that are networked together. However, the most significant source of fault current on the wind farm is the transmission system to which the farm normally ‚Äúbackfeeds‚ÄĚ power. Any transmission line capable of ‚Äúreceiving‚ÄĚ significant amounts of power from the farm is then capable of ‚Äúdelivering‚ÄĚ many times as much power should a fault occur. Catastrophic arc fault events have been the unfortunate result of a deadly mix of these astronomically available fault currents and novice electrical workers. Several years ago, one such fatal event helped initiate collaboration between our company and several wind turbine maintenance companies to develop electrical safe work practices training geared to the unique hazards associated with this rapidly growing industry.
Another common layout for towers is to locate a dry-type transformer ‚Äúup-tower‚ÄĚ and locate a medium-voltage SF6 gas-filled switch at the base of the tower. Having a large transformer up-tower in the nacelle creates serious working space issues. Additionally, setups like this necessitate long vertical drops of 34.5kV cable running in close proximity to the service ladder (Photo 2). It also means even more critical strain relief techniques for medium-voltage cables than are required for low-voltage cables.
Worker safety issues
When hiring turbine technicians, most wind turbine companies tend to focus on the prospective employee‚Äôs previous mechanical skills. Indeed, that is the skill set workers need more often for turbine maintenance. Although electrical troubleshooting and maintenance make up much less of a wind turbine technician‚Äôs routine work activities than you might think, even the most fundamental electrical troubleshooting is high risk. Despite its low frequency, the high severity of an electrical accident in this work environment produces a high risk.
As a result, lack of electrical background in this industry can be lethal. Here‚Äôs just one example. One inadequately qualified turbine tech was fatally burned by the plasma blast he created when he mistook a tap changer for a load break switch. He was in the process of de-energizing a 2MVA transformer for a routine maintenance shutdown when this horrific accident occurred.
In addition to NFPA 70E training, our company conducts arc flash hazard analyses for wind farms. If we hadn‚Äôt had the physical evidence from previous multiple wind farm accidents, we would probably have had a hard time believing the off-the-charts magnitude of the incident energy levels our software was spitting out (click here to see Photo 3A) and (click here to see Photo 3B). These studies have confirmed the obvious: Potential incident energy on most wind farms is extremely dangerous.
Code violations illustrated
The hazards and installation problems are not just confined to explosive levels of incident energy. As a retired electrical inspector, I see numerous NEC violations ‚Äúdown on the farm.‚ÄĚ One of the first questions I generally ask my contact person on a wind farm is, ‚ÄúWho is the AHJ responsible for inspecting your turbine and tower installations?‚ÄĚ More times than not, I get the same, strikingly consistent, shrug of the shoulders. In other words, they have no idea.
This raises another issue. Does NFPA 70 and 70E apply to wind farms? NFPA purists might contend that because many wind farms ‚Äúare on property owned or leased by the electric utility for the purpose of ‚Ä¶generation, transformation, transmission, ‚Ä¶ of electric energy,‚ÄĚ as stated in NFPA 70 and 70E 90.2(B)(5)c, these standards would not apply. Such purists would defer to the more performance-oriented National Electrical Safety Code (NESC) published by the IEEE. However, because the NFPA 70 and NFPA 70E are somewhat more prescriptive than the NESC, most wind farm companies have chosen to use all three standards. Ultimately, wind farms must comply with OSHA. Adopting NFPA‚Äôs consensus standards and IEEE‚Äôs NESC provides the best chance for OSHA compliance.
Unfortunately, enforcement of these standards can be lax. As a result of this apparent lack of oversight, I have encountered NEC violations that make me cringe. The most shocking offenses tend to do with egress requirements (Photo 4). On several tower designs, switchboard enclosure doors are hinged such that they open across the only exit path out of the tower. Having personally witnessed the excruciating death (not on a wind farm) of an electrical worker in large part because of inadequate egress, I‚Äôm especially vigilant about this violation of NEC requirement 110.26(C).
Another NEC violation I‚Äôve seen (Photo 5) is the intrusion into the work space [NEC 110.26(A)(1)] by containment walls for the step-up, oil-filled, pad-mounted transformers at the base of most tower configurations. A critical work procedure is performed in the 34.5kV side of the transformer ‚ÄĒ protective grounding. The darkly tinted arc flash hood worn during this operation makes the containment wall, if placed too close to the enclosure opening, an especially worrisome trip hazard.
DLO-type cable is commonly used throughout wind farm power systems. As shown in Photo 6 on page C34, there was no lug labeling at this location stating this 650-strand 262.6kcmil cable could be terminated on this particular breaker. This appears to be a violation of Secs. 110.3(B) and 110.14 of the NEC. A new sentence has been added to 110.14 in the 2011 edition of the NEC that states, ‚ÄúConnectors and terminals for conductors more finely stranded than Class B and Class C stranding as shown in Chapter 9, Table 10, shall be identified for the specific conductor class or classes.‚ÄĚ However, the 650-strand conductor shown in Photo 6 does not fit into any of these classes.
The conductors in Photo 7 are not grouped as required by 300.20(B) of the NEC. As noted in this section of the Code, ‚ÄúWhere a single conductor carrying alternating current passes through metal with magnetic properties, the inductive effect shall be minimized by (1) cutting slots in the metal between the individual holes through which the individual conductors pass or (2) passing all the conductors in the circuit through an insulating wall sufficiently large for all of the conductors of the circuit.‚ÄĚ By not having all the phases of a circuit grouped through a common opening in a ferrous metal enclosure, circulating currents will be induced in the metal enclosure. This could potentially produce enough heat to damage the conductor insulation where it passes through the hole.
What does the future hold?
One of the biggest contributing factors leading to this lack of oversight is the complexity of the business model for this industry. Much of the equipment is manufactured in other countries, which makes it non-compliant with U.S. standards. Also, because of the typically remote locations of wind farms, installation and tower construction are done without the benefit of building permits. Even if local jurisdictions chose to inspect such installations, it is doubtful they would have the technical experience to adequately enforce applicable codes.
Multiple layers of ownership and liability further complicate matters. In some instances, responsibility for worker safety falls to multiple employers. In fact, the responsible party for safety is often split between the tower owner, the contractor responsible for ‚Äúbalance of plant‚ÄĚ equipment, and the maintenance contractor who maintains the low-voltage equipment up to and including the turbines.
At one of my recent presentations on this subject, I found my audience was largely made up of insurance company representatives. During the Q&A exchange, they expressed a consensus of concern about how to insure this emerging sector, confirming my suspicions about a lack of oversight. They expressed a general sense of agreement that the wind industry‚Äôs chain of command and layered ownership make their risk/responsibility assessment difficult.
I tell the turbine technicians in my classes that I‚Äôve got good news and bad news. The bad news is you have elected to work in one of the most potentially dangerous electrical workplaces on the planet. The good news is if you take the proper precautions ‚ÄĒ and really understand the design of these systems ‚ÄĒ you can expect a long and worthy career ahead of you.
Kardon is a consultant with Code Check Institute in Philadelphia. He can be reached at firstname.lastname@example.org.