Think you’re paying too much per kWh?

Maybe you should consider starting your own utility.

Something like this came to pass late in 2005 when Toronto-based Barrick Goldstrike Mines Inc. powered-up its new 115.6 MW generating plant, built primarily to service the load of its operations northwest of Elko, NV. At the mine, two large milling works and assorted energy-intensive equipment steadily eat-up 125-plus MW of power. For local utility Sierra Pacific, this humming mountainside mine was pure gold, too— “one of those really beautiful loads that a utility likes to have as part of their portfolio,” observes energy consultant Ken Pavlich. It’s stable, big, and steady all year round.

By the same token, though, such an industrial power-gobbler can easily become captive to utility policies, which constrain market choices, politicize the rate-making, and effectively skim a hefty chunk of revenues. “The question becomes,” says Pavlich: “Are you really getting the kind of price from the utility that reflects what an important part of their service area you represent?”

Answer: In states where large industrial sites pay 3 1⁄2 cents or so per kW/hr, there’s no complaint; but Barrick’s gold mine—“a huge customer”—was paying over 7 cents in this booming, mineral-rich state. Nevada today (perhaps contrary to long-ago conditions) has established, he notes, “among the highest industrial rates in the country.” Besides setting high tariffs, Nevada regulators also impose tough rules for businesses that may seek to depart the grid and buy elsewhere.

So, four years ago—and after several years of working with regulators, lobbyists, and legislators to open the door to large-facility self-generation—a small team at Barrick began actively exploring new generation options. Ultimately, in 2003 Barrick contacted Pavlich, asking, as he recalls, “to assist Barrick’s very lean staff in figuring out what options might be available and feasible.” Then, as the discovery process moved forward and the self-generation idea gained momentum, the team grew to include significant outside legal counsel; they were critical to spearheading a reform of utility law and lobbying effort. Several large firms—Parsons Behle & Latimer of Salt Lake City, and some Washington DC firms—collaborated. In addition, Barrick also hired Power Engineers, of Boise, ID, and other consultants who were well versed in utility and transmission regulations (e.g., Brubaker Associates in St. Louis and EnergySource of Reno, NV). Ultimately, this assemblage helped foster fundamental change in Nevada’s industrial utility policy. For Barrick, this resulted in a measure of energy market autonomy worth millions to its bottom line. But the process of getting there, Pavlich observes, “was complicated ... and not for the faint of heart or light of wallet.”

Breaking Free
In Nevada, a utility customer may purchase generators and build a power plant onsite. Moreover, under a utility reform called AB 661, by satisfying certain relatively tough regulatory requirements, a power user can also gain direct access to the state’s transmission system. Achieving the latter puts a user in “the big leagues,” vastly expanding the range of energy options. For instance, such a facility may then buy power from cheaper sources hundreds of miles away; and it may even sell power on the same system or on the local grid. In effect, it has rights almost equivalent to those of utilities themselves.

To qualify, an applicant must agree to build significant generating assets within Nevada to help ease the state’s soaring energy burden. Second, says Pavlich, “You would have to become your own scheduling coordinator” and be willing to deal directly with the ISO/RTO (regional or independent transmission system operators)—in this case, the Western Electric Coordinating Counsel. And third, a facility must meet the law’s “single-largest-contingency” provision, which requires that whatever is the customer’s single biggest potential loss of power must be duplicated with backup—i.e., with costly redundancy. In addition, in Barrick’s particular case this meant investing in one or more standby generators, plus bus bars, transformers, and high-tension line connections.

Would The ROI Be Worth It?
Calculating payback for a normal power project would have been comparatively easy; here, though, says Pavlich, “When you start talking about transmitting power and purchasing market power,” along with building your own 100-MW-plus onsite power, “then the equations get a little more interesting and complicated.”

A suitable location for the mine's power source was found near Reno, NV, almost 200 miles away.

First, he explains, “You have to fully understand baseline load rates with the utility” and how these compare with alternatives. “And that’s the kicker,” he adds, “because the base load is often attractively rated. The utilities want very much to service all customers—and certainly all of them want to maintain these large, stable loads. So,” he says, “they’re going to work hard to try and keep their power rate to the industrial customer such that it never make sense for you to leave.” Thus, there’s some lowball pricing offered on the base load. The utility can eventually make up for this “discounting,” by heavily surcharging the ratepayer for exceeding the base and for peaking loads.

Another section of AB 661 allowing exit from the utility system provides that a company which meets all the just-noted qualification and others shall be granted a pro rata share of the transmission system on a “firm import” basis. “What that means,” he explains, “is that Barrick is granted 40 or 50 MW of import capacity,” i.e., access to potentially much cheaper power, which the utility may not interfere with. By having such access, Barrick could import enormous electricity resources, as needed, cost-effectively—even after installing its own onsite plant. During off-peak times, Pavlich notes, regional capacity surplus is such that “It makes sense to shut down some generating units” and buy from hydro plants in the Pacific Northwest or coal plants in Utah. “During those periods, Barrick might be running only 20 MW or 30 MW in the plant for load-following purposes,” he says, “and purchasing the rest off the market. They can do that literally on an hour-by-hour basis if the opportunity is there. They have the ability to do opportunistic purchasing of market power.”

Conversely, during the peak power times, when prices are highest, Barrick could supply its own electricity onsite, more cost-effectively—“maybe 100 MW from their plant,” he says. “And buy 35 MW on firm import.

“Attractively priced market power is frequently available,” he notes, “and transmission capacity in excess of their firm-import amount is accessible about 95% of the time. And so, in effect... what Barrick has is 165 MW of guaranteed available power, in order to service 125-MW to 135- MW.” He sums up: “They’re almost always purchasing significant amounts of power.”

Which is why—well before the plant was commissioned in December 2005—Barrick hired an energy management company, Avista Energy of Spokane, WA, “who is constantly in the market looking for prices for gas and prices for power,” he says. Avista also handles scheduling and coordination with the ISO and juggles the mine’s power needs. Avista checks market prices and decides, say, to operate the Barrick mine plant at only 20-MW minimal output, while purchasing 110 MW from the open market. “They’re putting sufficient power into the system,” he says, “so that the mine site is never leaning on the utility.”

Bottom line: All of this flexibility, freedom of choice and potential savings for Barrick stems from being proprietors of their own virtual power plant, meeting the terms of AB 661.

Siting at “Western 102”
Besides estimating the above payback curve, Pavlich and the other advisors also examined the logistical challenges posed by Goldstrike’s unusually remote locale.

The Helsinki, Finland-based Wärtsilä Corp. makes some of the world's largest industrial engines.

For one thing, no pipeline gas came near the mines. Also, the mine’s 6,000-foot elevation and frequently 100-plus-degree summer desert heat would significantly derate turbine engine performance. Finally, the cost and scarcity of water would impinge on the use of combined-cycle steam generation—the configuration that is more typical of large, efficient energy production. Pavlich observes: “Getting water permits and purchasing water are not trivial issues in Nevada.”

For some time initially, the Barrick development team still considered building a plant using LM6000 turbines in combined cycle; these are, pound-for-pound, relatively cheaper to run than are most other engine plants. With them, he adds, “You get a little better heat rating and a little better maintenance cost.”

But the killer would be the water usage, which would raise the operational cost substantially. And, again, there’s the turbines’ environmental derating. Turbines were now starting to look iffy.

In addition, the flexibility of a turbine-based combined cycle plant—even equipped with an aero-derivative model like the LM6000—was not likely to provide the easy access to opportunistic market purchasing that Barrick was hoping to achieve.

Getting permits for large diesels also would be a challenge.

Finally, he points out, “The other problem is, eventually the mine’s going to close—and do you really want a $100 million power plant sitting in such a remote location?”

Resolving these siting issues yielded this rather unconventional conclusion: The mine’s “onsite” generation didn’t really need to be sited at the mine at all. Rather, a more suitable location lay 200 miles away, near Reno—within a stone’s throw of the largest power plant that Sierra Pacific owns, the Tracy Generation Station. An adjacent parcel designated Western 102 was selected, he says, “simply for access to the available gas and transmission lines, as well as the fact that, in the future as the mine load decreases, that power plant would be very nicely located to service the load from the Reno, NV, area—which is the major load of Sierra Pacific Power Company.”

Adjacent transmission infrastructure would allow exporting the new plant’s power to the same grid, too. Thus, when the Goldstrike Mine eventually closes, its 100-plus MW plant can still go on—becoming, in effect, an important supplier and utility market competitor.

Atypical Engines For Unusual Conditions
Next issue: what gensets to buy and how many?

Since combined-cycle turbines were out, in this case, single-cycle reciprocating engines with the lowest possible heat-rate made the most sense. And recip engines, he notes, “don’t require water to attain the low heat rate, power output, and emissions guarantees, even in the hot, dry, high elevation.”

Also, flexibility in terms of operational characteristics “was very high on the list.” Pavlich is referring to the ability to turn resources on or off readily; to keep them on warm idle; and to do load-following rather than continuous running.

Another factor to weigh (again) was the single-largest-contingency rule requiring a backup for the biggest engine. This alone dictated that whichever engines were selected should be comparatively small for such a significant output. On the other hand, though, he observers, “You want the largest reciprocating engines available,” because the plant would output significant megawattage.

Flexibility would also be increased by having many smallish engines in parallel instead of a few very big ones.

In sum, everything logically pointed to a plant consisting of banks of reciprocating engines.

But which one?

Theoretically, plant designers could have installed dozens of 3-MW reciprocating engines, of which there are many good choices on the market; but this would have greatly multiplied the plant acreage and maintenance. “We got to the point,” he recalls, “when we realized we needed to find fairly large generation units,” i.e., in the 10-MW range. Here, though, the selection isn’t so great.

Giant Finnish Firm’s Big Recip Engine
Barrick’s engineering advisor, Power Engineers, knew that a Helsinki-based industrial concern, Wärtsilä Corp., makes some of the world’s largest internal combustion engines, primarily for marine propulsion. Although a big player in energizing decentralized plants abroad, Wärtsilä has built only a few in the US, such as the 111-MW Plains End facility in Denver. Nevertheless, its engine technology, Pavlich says, “turned out to have the best heat rate of any of the units that we looked at.” It’s “much, much better than most simple-cycle heat rates,” he adds. The rate is guaranteed at 8,700 (i.e., 8.7 mmbtu per 1 MWh. output). That’s about 20% higher than, say, a Frame 7 combined-cycle turbine. Of course, the lower the heat rate the better. But the difference in this case narrows considerably when factoring-in the environmental derating of the turbine, plus the added water expense. The Wärtsilä product, he adds, is “relatively insensitive” to high elevation and ambient desert conditions.

Total plant output is now guaranteed at a minimum of 115.6 MW—the largest of its kind in the US.

Dennis Finn, Wärtsilä North America Inc.’s business development manager, comments. The Wärtsilä 20V345G model, he says, uses a lean-burning fuel gas/air ratio and precombustion chamber “to attain the lowest heat rate in its engine class, along with very low emissions rates.” Moreover, he adds, when the engine runs at 50% load, the heat rate stays at about 88%—again, making it tops in its size and type class. Electronic controls maximize efficiency, even in the relatively narrow operating range of the lean-burn mode.

As for Wärtsilä’s water usage, it’s virtually nil. The engines’ air-cooling radiators on a closed loop allow the gensets to reach their performance specs with only two gallons per engine per week; this is needed for washing the turbocharger and maintaining the cooling loop.

On all counts, the Finnish product seemed a good fit.

Next question: How many to buy.

Per-engine output at site conditions (assuming 95°F ambient air and 4,400 feet above sea level) comes to 8,439 MW. The mine’s base load is 125 MW. So, in order to come close to this while leaving some margin for desirable electricity importation, the Barrick plant development team recommended installing 14 gensets.

Wärtsilä handled the project engineering and construction—providing one-stop shopping for both the schedule and performance.

Having multiple, flexible units enables the plant to have "both ready reserve and spinning reserve."

Commissioning took place in November 2005, just 14 months after the contract’s signing.

Total plant output is now guaranteed at a minimum of 115.6 MW—making the Barrick Goldstrike plant at Western 102 the largest internal-combustion gas-fired power plant in the US.

Operational Advantages In Multiple Engines
Plant design and operation are now fully maximized for flexibility—i.e., for rapid start-up and independent dispatchability. These are critical to attaining the full benefit of the genset configuration. For example—and a bit unusually—instead of employing one shared synchronizer, each engine is independently grid-synchronized to facilitate a 10-minute start-up from warm standby, and quick deployment. This qualifies the plant as ready reserve capacity for Nevada’s electrical grid, “which adds some income,” Finn notes.

Having multiple, flexible units enables “both ready reserve and spinning reserve,” Finn adds, as units can run on idle to qualify for the latter.

The plant can be cycled from shutdown to full load and back several times daily without negative impact. Many of the controls are automated, and fourteen employees staff the site.

One unit’s failure represents, he notes, just a 7% loss of full-load-rated output.

All in all, the configuration allows greater energy security, redundancy and continuity than would be possible with a couple of big turbines, Finn notes proudly, adding: “The plant is clearly fulfilling the need for intermediate and peaking power, and is qualifying for use as ready reserve.”

Dispatch of plant resources is done by Avista Energy, which has responsibility for monitoring the lowest-pricing option, hour by hour, i.e., producing watts onsite whenever this is cheaper, or powering-down and buying electricity on the market, whenever that’s advantageous.

As for emissions, selective catalytic reduction and oxidation catalysts provide control of NOx, CO and volatile organic compounds. Finn notes that early compliance-testing—such as grams/kWh NOx guarantee equivalent to 5 PPM at 15% O₂— is proving that Wärtsilä’s guarantees are being met without difficulty. Pavlich confirms: “It’s a very clean plant” meeting current best-available control technology.

It’s also well-positioned to continue operations for years to come, even after Goldstrike mining operations eventually cease. The adjacent Sierra Pacific plant is aged, with relatively costly production; hence, Goldstrike’s Western 102 plant stands ready to serve grid customers quite competitively.

Energy Independence Opportunities
How repeatable is this model?

Pavlich suggest that industrial sites with very big loads in high-electric-cost states could and probably should contemplate such an undertaking. But there’s a lot of overhead to weigh carefully. “To play the game that Barrick is playing,” he says, “which is, they’ve almost become their own utility, the economics have to work out right.”

If an energy-intensive industry can’t meet the high ante in Nevada or in other costly states, plan B might be to relocate to cheaper-power localities. Plenty of industries have already done so.

Pavlich observes: “We can look at many places around the country where lots of mid-sized industrial plants have picked up and moved, because power rates are too expensive.” Regions getting power from coal, hydro, and nuclear plants generally offer the lowest rates; conversely, states that face almost-certain escalating costs—which big energy users may want to consider vacating or avoiding—are those overly dependent on natural gas.

Meanwhile, utility commissions struggle to balance the competing demands of retail customers, industrial users, “NIMBY” environmentalists, air-quality constraints, evolving technologies, erratic fuel resources, and assorted market forces.

Pavlich Associates—many of whose clients are industrial customers looking for better energy options—is currently focusing on opportunities in the areas of heat recovery for power generation, and syngas, Pavlich notes.

Some clients are especially interested in shifting from reliance on pricey natural gas to synthesized coal gas.

In many ways, the ultimate goal of owning distributed energy resources, he says, is the ability to gain some measure of control “over what happens in the future” with a given industrial operation. A perceived “deep-pockets” ratepayer is ultimately at the mercy of the public utility commission. “Whenever a rate case is out there,” he observes, “you can take your best shot at arguing against it. But at the end of the day, you’re a captive customer—and you have to buy it.” Barrick Goldstrike is one of the fortunate few industrial sites that can now purchase many MW of power at true market prices. This ultra-flexibility is proving to be worth its weight in gold.

Pavlich sums up. “It’s a very innovative energy approach, and the entire team is very excited about what Barrick did and why. They took some risks, and it was not easy. The utility made it challenging for them to leave, coming up with more rules as the process was going along. But Barrick persevered and got the plant built—and they’re benefiting from it now, and they’ll continue to benefit for a long time to come.”

Based in La Mesa, CA, writer DAVID ENGLE specializes in construction-related topics.

DE - July/August 2006