The Marine Corps’ All-Out Offensive Against Wasted Energy and High Utility Costs

Twentynine Palms, CA, is home to more than 10,000 military personnel and their families who reside within a sprawling facility called the Marine Air-Ground Task Force Training Command. In relation to the Southern California Edison (SCE) power grid, it’s literally “at the end of the line.” Winters are mild, but summertime heat can hit 120°F and wreak havoc on the Marine Corps’ electric bills.

Combat units here receive training in combined air and ground maneuvers. Early in 2003, as United States forces were preparing for war abroad, Twentynine Palms’ utility managers were commissioning a Solar Turbines Taurus model for combined cooling, heating and power (CCHP). The key objective was to slice the facility’s reliance on grid power by about two-thirds. In short order, the Marines’ new minigrid would be tying together the output from this 7.2-MW dual-fueled turbine genset [making it “one of the largest cogen turbines” in the Department of Defense (DOD), notes energy manager Gary Morrissett], with the electricity produced by an onsite 1.2-MW solar photovoltaic (PV) array, together with two incoming grid lines from SCE. As payback, the Marines would be guaranteed savings in the range of several million dollars annually for the next 19 years.

During 2003, the first year of operation, Morrissett was able to scrutinize the turbine’s fuel consumption and power output on this highly automated system, using its built-in advanced computer controls. He found its efficiency could leap to “as high as 75%,” depending on the seasonal load – a figure about two and a half times more efficient than conventional power-company generation. In the low-load wintertime, Twentynine Palms draws from 6 to 9 MW of power, but in the roasting heat of summer, the figures virtually double, soaring to 15 or 20 MW – “mostly for air conditioning, of course,” Morrissett says.

The good news is that, as of early 2004, latter-stage extensions of the cogen project were on target for routing the searing exhaust of the big Taurus not only for the usual water heating but also for cooling power, by linking to three new and/or upgraded thermally activated absorption-chiller plants.

One key strategic rationale behind this CCHP enterprise, at least initially, was the fact of Twentynine Palms’ remoteness relative to utility-company generating and pumping stations. The base’s commanders have long had concerns about energy security. Backup generators are in place, but they’re not sufficient to handle the consequences of an earthquake or some other major interruption. Energy security “was a primary consideration to us,” Morrissett recalls. Another was the region’s notoriously high power rates, which probably made an investment in cogen not only attractive but almost inevitable, even for a public-sector site, given the potential for saving tens of millions of dollars, long-term.

The turbine idea had been kicked around since 2000, when Southern California Gas offered to install a genset for the military under a utility-service-provider (USP) arrangement. This would have been administered by the Marine Corps’ Southwest Division based in San Diego, CA. An initial feasibility study and design then were commissioned by the US Department of Energy’s (DOE) Office of Distributed Energy Resources (very keen on cogen). SCE then produced the requested detailed rate and tariff data. Out of all the number-crunching emerged an extremely attractive payback curve, along with a few challenges, such as natural-gas supply and even the adequacy of line pressure to pump it into the base.

As the assessment went on, the concept of doing the project under a government-designed Energy Performance Savings Contract (ESPC) emerged. ESPCs basically enable federal agencies to hire contractors to help them reduce their energy bills and then share in the net savings. Such contracts are often more cost-effective and easier to administer for government managers than a USP relationship. With an ESPC, says Morrissett, “you’re guaranteed savings,” and, should they not materialize, “the difference is made up from the contractor’s portion. So there’s very little risk on the government end.” Too, contracts can put the burden for maintenance, repair, and replacement on the vendor, providing a strong incentive not only for correctly assessing the potential savings, designing and building the system, but also for its continued high efficiency.

The Marines solicited an ESPC contractor and in September 2001 inked a deal with Johnson Controls Inc. (JCI). For the next 19 years, the base would be guaranteed the production of about 55 million kWh/yr. of cogen power, thereby reducing its electric bill by $5.8 million/yr. All in all, Morrissett’s department would be looking at “something like $57 million in total that we could work with,” during the contract’s duration.  All of that money could readily fund more energy projects under the same ESPC plan – and additional savings also would be guaranteed.

Another attractive provision was that JCI would handle the technically complex challenges of reliably projecting the savings goals, specifying appropriate products and soliciting bids, and completing and implementing the final design, always keeping Morrissett in the loop.

Funding was arranged through the US Army Corps of Engineers and the Naval Facilities Engineering Service Center.

“One of the most interesting aspects is how the project came together,” says BP Solar’s Mac Moore. “Solar energy is of course our specialty, but here it is just one of a suite of onsite generating and efficiency measures that the project manager, Johnson Controls, had to work with in developing a total solution for the base.” Moore believes that this approach in conjunction with the use of ESPCs paves the way for innovative financing of future projects.

Groundbreaking took place in May 2002. West Coast Air did the major installation and was joined by Walsh Engineering for the design and Baker Electric for the wiring. A 3-mi.-long high-pressure natural-gas line was laid to ensure the turbine’s fuel supply. Stage one – installation of the big Taurus inside a 7,200-ft.² turbine hall – then was completed within about half a year, and the new system was fully commissioned in early February 2003. The total cost, including all fittings and a grid connection, was $16.2 million.

CHP plant under construction: lowering the turbine into the turbine hall

Running 24/7 (except for two weeks yearly of maintenance downtime) the turbine burns about 50,000 therms of natural gas monthly. Although rated for 7.3 MW, the output actually comes to around 7 MW, due to combustion inefficiencies at higher elevations. Fuel consumption and power output also are turned down slightly in the lowest-load winter periods, in order to maximize efficiency.

After yielding up its electricity, the turbine’s hot exhaust is captured in a heat-recovery generator to supply heat for roughly a third of the base’s central heating plant; this consists of three International boilers outputting 40 million Btu/hr. apiece. The Taurus actually was sized “in order to supply heat for one of those boilers,” says Morrissett, since boiler load is a prime scaling factor in many cogen projects. “Whatever amount of [exhaust] heat you can use, and not waste your gas producing it,” he says, will tend to determine the appropriate turbine. During high desert winters, which are sometimes mild and short, two boilers typically might be running, but in summertime, their gas valves are closed because the base’s hot-water needs can be supplied largely by the Taurus’ exhaust – to the tune of 3035 MBtu/hr. – as planned.

This hot-water output then is pumped through about 40 or 50 mi. of piping, at up to 2,400 gal./min., to supply 80% of Marine-base buildings “with very-high-temperature hot water for domestic use, some steam applications, and building heating,” notes Morrissett.

All of the boilers and the Taurus genset can be fired with either natural gas or diesel as a backup.

In sum, the system easily is attaining its originally targeted efficiencies and ensuring greater energy security. “We’re very happy with the cogen unit,” Morrissett says. For 2003, it reduced the wattage formerly being purchased from SCE by two-thirds. Natural-gas purchases shot up from the preinstallation rate of 150,000 therms to more than 500,000 and will remain high indefinitely. If this setup were running in certain private-sector locales, this huge surge in projected consumption might have been a deal-killer due to volatile natural-gas prices, but the feds can buy anything they need in bulk through aggressive purchasing: The fuel-cost impact isn’t quite so critical. Even with higher-priced natural gas, and with the cost of system maintenance added, the first-year savings will come to an estimated $5.8 million – which again is being divvied up by the Marines and JCI.

To ensure that these enormous savings keep rolling in, an elaborate monitoring and verification system integrates and logs the natural-gas input and electrical output meters, together with 10 other critical variables. Operators can access all of this data in real time via a fiber-optic communication cabling, which connects the plant with monitoring and control at a nearby substation where the power enters the base.

View of turbine hall with black-start generator outside building. The CHP system is on-line.

The entire $16.2 million total outlay easily could be paid off at the present “income” rate in less than four years, but instead of doing so, Morrissett and other base managers decided at the outset to begin plowing their share of the dollars into buying more energy upgrades and assorted other related goodies, such as several hundred four-by-four daylighting and skylighting panels so far. Tops on the Marine Corps’ wish list of fully funded major investments were the following:

PV was a natural choice for a desert facility; the Mojave enjoys 320-plus sunny days a year, and what better way to offset the peak-rate charges from SCE? Too, both DOE and DOD want to see government facilities converting to more renewable energy sources and in fact are hoping to achieve a shift of 10% or so by the decade’s end. “This is not an order but a recommendation,” Morrissett points out.  Also enticing the Marines were generous incentive funds – amounting to $0.045/W of PV power production – offered by the California Public Utilities Commission. This, combined with the preapproved funding the Marines were getting through the ESPC savings, made the PV investment a no-brainer. Managers at Twentynine Palms opted to build “the largest system we could” with these funds, he recalls. Two firms, BP Solar and Powerlight, were selected as contractors; groundbreaking occurred in January 2003, and the system was operational by September. The resulting array includes 8,706 solar panels, each capable of yielding 150 W, or 1.2 MW total. The system uses single-axis tracking, which feeds into five inverters and then into five transformers and into a main line that carries it into the cogen plant.

Overall, Morrissett says, the PV is performing “fairly well,” although “there are times when one of the 15 separate tracking systems misaligns. But this is an easy fix.” Due to assorted limiting factors, the actual PV yield has averaged around 900 kW. During its first summer, this amounted to about 5% of Twentynine Palms’ total electric needs, with the Taurus contributing nearly half and the SCE grid contributing all of the rest.

This wasn’t part of the original cogen blueprint in 2001, but the idea of installing big, high-performance air conditioners (AC) sprang to mind very quickly when Morrissett was eyeing the base’s rather shaky 15-year-old chillers and wondering how to spend the Marine Corps’ share of the ESPC dollars. Moreover, he adds, “Once we got the cogen in place and saw exactly what the loads were going to be, we knew that we wanted to go to a central chilling plant – so we ended up incorporating the absorbers in there.” New chillers, running full-blast in the summertime, also could utilize much of the turbine exhaust and thereby achieve even higher efficiencies.

In the summer of 2003, the turbine exhaust began powering one 200-ton absorption chiller for turbine inlet air cooling (which makes it run more efficiently), in addition to providing all of the base’s piped hot water. Ambitious plans are already well underway for much more chilling. As of the winter of 2003-2004, one big new chiller plant was nearly finished; construction of two more to house multiple high-efficiency Trane R123 chillers was also on the near-term horizon. When upgrades to the four centralized plants are finished in mid-2004, they’ll be capable of yielding a combined 2,600 tons of cooling.  A chilled-water loop will circulate to output to around 40 or so buildings on the base, thereby replacing 80% or more of the aging package units.

Chilled-water lines also are being laid out with an eye to servicing future base structures so when the latter eventually are built, Morrissett says, “we’re just going to be tying into those lines so we won’t have all of the complex mechanical AC systems going in, with those buildings,” thus saving even more on up-front and operational costs.

Yet another application comes from this trigen coolant: A small chilled-water loop now is icing down the refrigeration condensers in the newest mess halls. Using water-cooled condenser coils “hasn’t been done before to my knowledge,” he believes, but at least in concept, it should afford the coils higher cooling efficiency and longer life. Come winter, when the main chilled loop is shut down, this small spur into the chow halls will continue to circulate under its own pump; the main chilled-water lines then will act like a heat sink, he adds. Thus far, this innovative system “seems to be working out fairly well.”

To be or not to be grid-connected? And if so, how?

Both of these are thorny questions for cogen designers, particularly when turbines produce surplus power that ostensibly could be sold back to the power company to yield the owner a bit of income through net metering. In the summertime, this would be irrelevant at Twentynine Palms because the base needs all of its own power, but during winter, the facility has an onsite capacity for generating around 8 MW all told (i.e., combining the genset and PV systems); the electrical load needed hovers between 6 and 9 MW. California has a net metering threshold of 1 MW. This means that, at least in theory, the Marines might be able to eke forth some occasional small excess of its self-made power to put back into the SCE grid. JCI engineers actually weighed this option in their design and finance calculations. Ultimately, though, it was decided that the potential income wouldn’t be enough to offset the additional expenses and headaches of coping with the issues necessitated by net metering. “We would have liked to export power,” Morrissett recalls, “but because of the interconnection agreement and what Edison’s concerns were at the time, I think it was decided that we should not.”

Interconnections, he adds wryly, “are interesting.'” Any utility, SCE included, requires that rigorous technical controls be applied to would-be grid connectors “whether you are going to export power or not,” he says. This rigor is ostensibly required to ensure that a customer doesn’t export power inadvertently. As for the Twentynine Palms grid connection, “even without net metering, our agreement was a challenge to get through.” Negotiations with SCE were handled by government contract managers from the Marines’ Southwestern Division. This entailed a complex four-way tussle among the Division, Twentynine Palms’ managers, JCI, and the utility company. Under the resulting agreement, the base was required to receive a small amount of energy coming in at all times, he continues, “and just because of the load swings on the base, we set that at about 800 kilowatts.” An added – albeit minor – complication was Twentynine Palms’ combined cogen and PV output. “We were forward thinking on this,” says Morrissett, “and while the cogen system was under construction, we went ahead and put in an extra breaker for the PV system we knew we were going to build. The PV system actually ties into the breakers at the cogen plant.” Thus, “all of the controls in the cogen plant actually control the PV system as well.”

SCE’s grid enters the base from a 34.5-kV main and a 12.47-kV secondary feed, serving one section of buildings. The cogen and PV feed into the latter via four main breakers. This separation of the two subsystems is what basically prevents the cogen power from leaking back into the grid. “There are times,” says Morrissett, “when we’re actually overgenerating to those four main breakers and putting out power to the rest of the base” (i.e., to areas served by the main power line; however, a monitoring station prevents any current from flowing back outside the base).

“It is a very complex system,” he decides. Several computerized control panels link the elements with lots of wiring to make it run automatically. The combination of two incoming grid lines, a very big genset, and a field full of PV cells, all powering multiple heat and cooling applications, “just makes it quite complicated,” he says.

Morrissett has nothing but praise for JCI for designing it and making all of the parts work (so far) almost flawlessly. Long-term maintenance by JCI is part of the deal; recently the contractor agreed to assign three full-time maintenance techs to Twentynine Palms to make sure the components keep running smoothly. Says Morrissett, “We’re very happy with Johnson Controls.”

Complicated though it is, this particular installation is also, so far, a resounding technical and financial success. It dramatically illustrates how cogen power can not only pay for itself readily but, with the sometimes-considerable savings that result in larger projects, also underwrite additional energy measures – thereby leveraging the initial cogen investment even more.

Regarding their initial goal of increased energy security, the Marines have the situation well in hand, with some 8 MW of dual-fueled genset and solar onsite power.

As cogen demonstration projects go, Twentynine Palms ranks as one of the largest and most efficient examples of CCHP cogeneration and thermally activated systems anywhere, particularly in the public sector. DOE’s Federal Energy Management Program wants to accomplish similar savings at facilities worldwide and points to Twentynine Palms as a showcase project. It also underscores the importance of fully incorporating thermally activated system opportunities as well as heat-energy recycling in the cost-benefit analysis. Morrissett concludes, “We’re very happy with the project.”