With rolling blackouts and unbelievable electric bills, California's energy fiasco four years back was front-page news. It also made any investment in standby local power look sensible. Business owners perceived the added reliability and affordability of onsite generation as tantamount to their very survival. Out came the Excel spreadsheets, and number-crunchers gleefully began figuring how quickly—a year? two? three at the outside?—a generator would pay for itself against those insanely overpriced kilowatts.

Then, suddenly, in late 2000 the cost of natural gas—the fuel for most onsite power cogeneration—shot up, too. All those rosy return-on-investment figures were out the window.

Fortunately, by 2001 the State of California had intervened, and prices began drifting back to Earth. But utility customers were left with an obligation to pay for future energy contracts signed at the peak of the market run-up; some had locked-in rates as high as 10 times the current (i.e., late 2004) kilowatt price.

Having been stung badly in 2000–2001, many business owners began fetching around for an escape from this sole-supplier dependency. One San Diego commercial property firm, Collins Development Company, sought advice from an engineering partnership in Orange County, Craig Hofferber and Anton Paley of H&P Systems Inc., specialists in distributed power generation and HVAC solutions for buildings. How, asked Collins' Asset Management Group, could an office building owner save money with combined heating and power on-site? Collins knew, too, that some financial industry tenants might gladly pay a premium for the security of backup power in the building.

By 2002, H&P already had been involved with more than a dozen onsite power projects, Paley recalls. They'd also performed numerous energy-related assessment surveys to determine viability. In doing a basic screening of this kind, he points out, "Step one is simply to determine that, if a generator were installed, the savings on combined gas and electric bills would recoup the investment in hardware in a reasonable time."

Such a screening typically involves researching historic electric loads in the intended facility; evaluating the compatibility of existing HVAC hardware; looking at thermal load possibilities; gauging electric rates at various times of use; taking all of this to do a thumbnail assessment of how much electricity might be saved with onsite power; and, finally, determining the answer to a key question: "What is the existing cost of heat energy used at the site?" Paley explains.

The latter refers to the currently used volume of natural gas that is burned to heat water and warm tenant workspaces, or to provide process heat in an industrial site, or even for use in absorption chillers for office air conditioning. However it's used, heat utilization is critical to any combined heat and power (CHP) cogeneration, because this volume of usage will instead be piped to an electric generator first; the resulting exhaust of the generator will be captured to heat up the hot-water systems. The resulting electricity thus becomes "free" in the sense that this same fuel volume was already being expended for heating alone.

In addition, by making sure to run the generator during times of peak electricity rates, the relative savings rise even more. Eventually, the combined economizing should recoup the cost of the generator and assorted hardware. At the same time—and as a not-inconsiderable bonus—the generator offers peace of mind against the vagaries of brownouts or grid delivery failures. In 2002, this was a big concern. It still is, Paley notes, primarily because the peak demand on the southern California power grid occasionally approaches the peak generating capacity being distributed to the utilities.

Four Stories, Two Custom Retrofits

Of the six buildings that Hofferber and Paley surveyed in 2001, four turned out to have relatively low electric loads, and their rooftop packaged-unit air conditioners didn't use centralized chilled water, making them incompatible with a viable cogenerator. That left two others having the right ingredients for CHP: One, the four-story, 35,000-square-foot Citibank West FSB on Herschel Avenue in the coastal city of La Jolla; and the second, around the corner on Ivanhoe Avenue, the Wall Street Building, also four stories tall and having approximately 30,000 square feet of floor space. Both structures happened to house similar tenants—a mixture of financial-service firms, investment brokers, attorneys and other upscale professional offices.

H&P does power designs for a variety of places, but wherever a building needs less than 200 kilowatts, the engineers often turn to a 60-kilowatt natural gas microturbine as the best solution. Paley finds the Capstone C60 readily scalable to match the thermal and electrical load profiles closely. Two, three, or more units can be arrayed together in an efficient chain; they require little maintenance; they are mechanically reliable, offer high combined fuel efficiency, emit low Nox, and are quiet. All of these virtues are critical in office or residential power plants. H&P has recommended or specified C60s in a dozen or more projects to date.

At the Citibank building, a Capstone C60 utilizes two hot-water-fired absoprtion chillers for air conditioning and the reheat loop, for interior comfort control.

The initial feasibility survey found that the Citibank West site could be well-served by one C60, utilizing two hot-water-fired absorption chillers for air conditioning and for the reheat loop.

The nearby Wall Street Building's thermal load (based on its air-conditioning need) was high enough to require 180 to 200 kilowatts of power. For this, H&P specified two C60s and a direct-fired absorption chiller. The site would not be able to use the turbine's exhaust heat for outright heating purposes, however, due to the HVAC system's design and "the cost to recover that marginal capital expenditure," Paley says.

At a price of just under $100,000 per Capstone C60—plus the installation and assorted equipment expenses—the total investment for a one-unit installation would come to around $200,000. Nearby, the Wall Street site, having two C60s, would cost another $100,000-plus.

The Challenge to Cost-Justify Cogen

Those dollar figures are typical of the savings that must be attained with any C60-based installation. Besides recovering that outlay in a reasonable time frame, the system must also be viable in light of erratic natural gas prices. A highly refined estimate of monthly operating costs must therefore be calculated into the investment projections, spanning several years, and it must incorporate rather painstaking estimates of negotiated natural gas pricing—all in order to justify the total upfront cost.

As for estimating that fuel usage and per-therm costs, weather and regional stores of natural gas will greatly effect those figures. Weather trends alone can lengthen or shorten the payback of a cogen project by many months. In this case, La Jolla's balmy coastal climate, which varies modestly from season to season, Paley notes, tends to moderate estimates. Energy users aren't struggling to pay for high heating bills, as those in the Northeast do, or facing long months of summertime air conditioning. Instead, they're tempted to pull the plug on their HVAC systems and open the windows. How could cogeneration be viable there?

The answer is that, notwithstanding their climate advantages, La Jolla and adjacent San Diego in 2001 and 2002 were still paying exorbitant electricity bills, and that, along with fears of unreliable power, shifted the assessment in the project's favor. Although natural gas prices were high, this particular region also relies on natural gas for some of its power-generation plants. As a result, fuel prices paid by the utility should rise or fall in tandem with those of smaller onsite power generators, making an investment more sensible than it might be elsewhere.

Too, there's that previously mentioned concept of virtually free power. The more energy that's needed for heating of any kind, the more of this bargain-priced electricity is produced. Hence, in cost-justifying a CHP investment, the critical factor is typically (again) gas utilization. In these two applications, heat exhaust would activate the buildings' absorption chillers. In both the Citibank West and Wall Street buildings, heat demand turned out to be the deciding issue. As Paley recounts, H&P designed the new system so that during the winter, "the turbine's exhaust heat would pre-heat the hot water going into the boiler—to heat the temperature control zones inside the building—and during the summer it would send the heated water to the absorption chiller." Even a seaside resort needs some air conditioning in office buildings to offset solar and internal heat gains, he adds. And in these two buildings, "the windows can't be opened!"

Trigeneration of this sort—combined cooling, heating, and power for buildings (abbreviated as BCHP and sometimes CCHP)—has become a more feasible option recently in small to medium-sized office buildings thanks to the improved reliability of exhaust-heat-fired absorption chillers. Not long ago, Paley points out, there simply was no such chiller small enough to mate cost-effectively with a microturbine. Absorption chillers tended to be large, inefficient, and unsuited to smaller applications. Several manufacturers have now downsized their chillers, and in particular, Paley and Hofferber have found that the Yazaki water-fired absorption chillers are a good complement to the Capstone C60. Yazakis can efficiently capture turbine exhaust heat to produce 20 tons of cooling—just about right for a wide range of buildings and, in particular, for Citibank West tenants. A similar pairing of absorption cooling and electric generation suited the needs of the Wall Street Building. There, a direct-fired waste-heat absorption chiller made by Broad was specified for its 80-ton output.

Newer, reduced-size chillers are opening up an important niche for CHP in mid-sized offices, Paley adds. Almost all of H&P's projects are now designed for high-efficiency trigeneration. This is particularly true in California, he says, where "it really doesn't make sense to install distributed generation just for baseloading, unless you go with combined chilling, heating, and power." However, he advises, because these pocket chillers' reliability can make or break a project, a would-be user should look for suppliers who can provide good local support. Most systems are currently being imported. In Paley's experience, the Yazaki and Broad models have proven reliable, but this can't be said of all brands.

When It's Payback Time…

Now comes the question of how quickly that $200,000-plus outlay will be repaid in net energy savings. In California the arithmetic hinges significantly on potential rebates in the range of five figures, made available through the California Energy Commission (CEC). To qualify, a system must utilize at least 42.5% of the heat generated by the engine in some energy-efficient fashion (i.e., "It can't go up the stack," Paley says). The CEC subsidy gives a generous 30% of the project cost, capped at $30,000. If the rebate gets approved, then a break-even point comes at around $170,000 instead of $200,000. Amortized over 60 months or so, that comes to about $2,800 per month, or $140 per business day.

Can Onsite Power Save That Much?

At this point in the investment research, it's common to seek out detailed electricity rate data, even down to tiny half-hourly or 15-minute increments. Such refinements help to cost-justify the system for peak shaving and peak lopping strategies (i.e., running the generator to coincide with high electric rates). In 2001 this level of refinement wasn't necessary, simply because monthly bills were all high, and any cogeneration investment looked pretty good. Under more usual conditions, though, it's important to analyze loads incrementally and in detail. Paley advises that an analyst should also match the load level carefully against tenant occupancy rates month to month. Otherwise, estimating errors might easily creep in, causing a failure to meet the investment-return criteria the owner expects.

When making their usual calculations, H&P and other developers apply accepted engineering and accounting measures; on the La Jolla job, Paley also confirmed his numbers with software provided by Capstone, which also ensured that Paley's designs would qualify for the CEC rebate. Similar software is available from other suppliers, but Capstone‘s version "accommodates all of the variables to be considered when selecting a site for application," he says, "and this makes for a great second opinion on the viability of a particular project scenario."

More Engineering Design Decisions

As it was polishing the spreadsheet numbers, H&P was also exploring a host of technical issues regarding wiring and electrical distribution. For example, it had to resolve the question of exactly how to supply both buildings with power, heating, and cooling. The respective wiring schematics differed significantly; thus, customized electrical designs had to be drawn for each. Citibank West sported an array of about two dozen small distribution panels routing the electricity building-wide, so it wasn't at all suited for having its C60 wired as emergency backup power: the access points were too numerous and diffused. On the other hand, the Wall Street Building had only a few large power panels; if a Capstone were wired to just two of these, the microturbine could supply virtually the whole building with adequate backup power, should the grid go down. Installing automatic transfer switches was fairly simple and relatively inexpensive, Paley adds.

As for daily operation, both systems were planned to rev up at about 6:00 a.m. and run until about 7:00 p.m. At Citibank West, automated turbine controls were integrated into the existing Johnson Controls Metasys DDC controller, which was running the building's HVAC system. The Wall Street building lacked automated controls, so the system was equipped with a time clock.

Paley, summing up the electrical concept, says that both buildings' generators were designed to run baseloaded in parallel with the utility grid. Sophisticated relay-protection gear would ensure that the attached grid would be secure from a power surge. "When our generators come on, they parallel in phase and in synch with the power being provided by SDG&E [San Diego Gas and Electric] to the building," he says.

Engineering drawings were completed by the firm that eventually built both projects, California Power Partners of San Diego. Paley suggested Collins hire Cal Power on a design-build basis—i.e., waiving the formal (and usually more expensive) bidding process, in favor of a negotiated fee basis.

Installation of the solo C60 for Citibank West got under way in the spring of 2002, and commissioning came in December. The total time—from the initial building assessments to full operation of the project—came to eight months. In the interim came a refined cost assessment; engineering design; some downtime while awaiting the rebate approval; demolition and removal of an existing incinerator; and finally, turnkey installation of the microturbine, the heat-recovery hardware, and the Yazaki auxiliary items.

Total price: about $202,500.

What About Cost Recovery?

H&P specified two C60s and a direct-fired Broad absorption chiller to meet the Wall Street building's thermal load, high enough to require 180 to 200 kilowatts of power.

Alas, since 2002, price gyrations have dominated the story. Even as the Citibank generator was being put in, electric rates at SDG&E continued to decline, albeit deceptively: the rates did not yet reflect the cost of those impending multibillion-dollar contracts signed by former Governor Gray Davis in 2000. Eventually those will come due. Rates will surely reflect whatever least-painful solution is hammered out by regulators.

In any event, the originally projected payback curve—which had been calculated at just three years—was now drifting way off. During 2003 and 2004, the curve stretched to four and a half years, then to six years. Naturally, this has come as a disappointment to the owners.

Further clouding the foreseeable impact and business analysis is the fact that Collins Development also decided to press ahead and install multiple efficiency upgrades: not only did Collins want onsite power and high-tech chillers, it also bought an array of related energy-efficiency upgrades for both buildings. The Wall Street site got, for example, a new, high-efficiency, energy-saving Trane chiller, which will run on its own conventional natural gas heat source for the evaporation cycle, as a supplement to the primary chiller—which is the Broad unit powered by the turbine exhaust heat. Replacing an older, inefficient chiller with these two high-efficiency models—only one of which runs off the turbine exhaust—made the most sense, because it allowed greater flexibility and created, in effect, a two-way backup option.

In addition to being outfitted with those chillers, both buildings were equipped with variable-speed drives for the air-handler motors, as well as energy-saving motors throughout, and an assortment of other upgrades, notes Andy Fraser, the building superintendent and head of maintenance for both locations.

Fraser oversees the systems' daily operation, and, as of mid-2004, he reported he was already noticing a "substantial improvement" in the operational efficiency of both buildings, thanks to the upgrades. Moreover, he adds, "The turbines are doing what they're supposed to be doing" (i.e., their exhaust heat is indeed supplying enough energy to the absorption chillers to cool the buildings on summer days). This has reduced the need for secondary air-conditioning support by about 50%, he estimates. Both the Broad and the Yazaki chillers "seem to be doing the job," he says.

Notwithstanding the business considerations involved, Fraser finds that it's fun having a microturbine to marvel at. Even after two years, there's still plenty of novelty and admiration being discovered by local architects, engineers, and maintenance people, who occasionally stop in to see it working. "They seem very impressed," he says. Better yet, "The owners are really excited about it, too."

As for quantifying the efficiency impact, Fraser thinks this won't be doable for another two or three years, and isolating the overlapping effects of those multiple retrofits at the two buildings won't be easy.

Paley, for his part, estimates that as of mid-2004, but prior to the summer cooling season, the Citibank turbine was probably lopping-off between $1,600 and $1,800 per month on SDG&E billings.

In 2003, Collins gave the go-ahead to install the two C60s at the Wall Street Building; in June 2004 they were slowly being commissioned.

Future Cost-Recovery Scenarios

In his screening method for current and future trigen projects, Paley is now applying a more price-sensitive assessment method, one more reflective of the wild up-and-down energy market, "looking at the dynamics of what gas pricing might do over a five- or 10-year period," he says. His cost-justification strategy when pitching trigen to new clients is more conservative; he tends to tell a site owner that if costs can be recovered within five or six years, the owner is still home free, and has made a good return. Capstone microturbines usually last 15 years or so, he notes, and so even if the fuel costs stay high, payback is eventually ensured. After the initial outlay has been recouped, he says, "The rest is gravy. You're operating in a fairly acceptable payback range with this investment." He adds that if a project is designed to match the heat load, "you really can't lose money."

As if to underscore Paley's point, at about the time that the second and third C60s were being fired up at the Wall Street building, in early summer, a local news story reported that SDG&E was being pressured to raise its utility rates substantially, especially for business customers, an estimated $500 more per month, for an average business like a restaurant or gas station. Bigger customers like the Wall Street and Citibank buildings would obviously pay more.

Moreover, just as that bad news was looming, SDG&E also was receiving approval for a billion-dollar-plus investment in new power plants, which obviously will be borne by local ratepayers as well.

All of this is now making Collins' decision, back in 2002, to invest in cogen projects for La Jolla seem prescient. If and when, a few years hence, big rate increases do arrive—as seems inevitable—this will almost certainly require a redrawing of the project payback curve again—only this time, much closer to the original expectations.

It all likelihood it's not possible to guarantee a rate of return on any power investment, Paley observes. But those in the market should probably consider this: Where do you think electric rates will be heading in the next half-dozen years?

In California, at least—where the bill for past mistakes surely will be coming due—the trend is definitely up.

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

DE - September/October 2004