"Campus Heat and Power," Feature Article, December 2006

campus heat and power

One university decides to practice what it teaches about thermodynamics.

by Lee Langston

it all started with a helpful suggestion. In 1993, I wrote a letter to the president of the University of Connecticut, where I am a professor of mechanical engineering, laying out what I thought was a pretty straightforward proposal. "One can imagine that it is a rare event for you to receive a letter from a faculty member proposing to substantially reduce an operating expense of the university," I wrote. "I am doing so here, and will even predict that the enactment of my proposal might make money for the university."

In the rest of the letter I briefly outlined a plan to build a cogeneration plant at UConn that would help to lower its yearly energy bill. Cogeneration, also called combined heat and power, is the production of more than one useful form of energy—both heat and electric power, say—from a single energy source, such as the burning of natural gas or some other fuel. In cogeneration parlance, a compact college campus is a perfect "host." A cogeneration plant, with its double use of costly fuel, must be collocated with the host, for acceptance of both electric power and heat output. Heat, either as stored internal energy (as in hot steam) or in transit by virtue of a temperature difference, cannot be transferred over long distances.

This is one of the three 7.5 MW gas turbines at the heart of the UConn cogeneration plant, as seen through the doors of the sound- and fire-proof enclosure.

Specifically, UConn's Storrs campus of 15,000 students already had its dormitories, classrooms, laboratories, and other facilities heated by steam generated in central plant, natural gas-fired boilers. During warmer months, campus buildings were cooled by individual electric-powered air conditioning units or by central plant chilled water, cooled by central plant refrigerant units. All campus electric power was purchased from the local regulated electric utility company and dispatched to the rural Storrs campus via a dedicated power line.

Southern New England utility electric power is mostly generated in a nuclear reactor and gas turbine combined-cycle plants, and is notoriously expensive. All in all, it seemed like an ideal place to try cogeneration.

Anyway, I quickly found that my letter made an impact, although not the one I expected. Shortly after I wrote my letter, I got a call from a vice president at our local utility company, asking me to go out to lunch. At our luncheon, he worked hard to convince me that my cogeneration proposal was perhaps not a good idea. It was not too long after that lunch that the utility company reduced the electricity rate charged to the university by about 10 percent. So much for my cogeneration proposal.

Shortly thereafter, the State of Connecticut made a decision to invest $1 billion over 10 years in the university's infrastructure. With $100 million in new buildings coming online each year, it didn't take long for the university's energy costs to shoot up, from $8 million in fiscal year 1993 to $15 million in 2002. By 2002, the university administration came to the same conclusion I had reached: A cogeneration plant for UConn made sense, especially with electric utility deregulation taking place and with, as a consequence, the unit cost of electricity to customers going up, not down.

That cogeneration system is now up and running. If it lives up to its early promise, it will prove to be a smashing success.

thermodynamic morality

My first exposure to a running cogeneration plant was a tour of a 300-megawatt facility in Rotterdam in 1988. The plant supplied the Dutch city with electrical power from generators driven by three large gas turbines (also called combustion turbines), each fueled by North Sea natural gas. The gas turbine exhaust gases passed through three very large boilers, or heat recovery steam generators, which produced steam used to heat Rotterdam's houses, offices, and other buildings. During warmer months, the steam, which was no longer needed for heating, powered a steam turbine to generate electricity, making the facility in summertime a combined-cycle (Brayton and Rankine) plant.

I was impressed with what I saw there and thought it made sense to emulate for a variety of reasons. It's not difficult to see that cogeneration reduces both the use of a costly fuel and the effect of its combustion products on the environment. On the very firm moral ground provided by thermodynamics—the nearest subject to a religion that many engineers have—one can say that heating buildings and people by simply combusting a limited, high-grade fuel to produce a low-grade form of energy is wasteful, and waste is the gravest sin in thermodynamics.

Cogeneration takes the thermodynamic moral high ground by using the fuel to first produce high-grade energy—electricity, which is readily converted into various forms of work, like turning an electric motor—and then, low-grade heat, with its lower values of thermodynamic availability.

The cogeneration plant was designed to blend seamlessly into the campus landscape. In fact, the doors shown here lead to a classroom.

Frequently, cogeneration plants are characterized as achieving thermal efficiencies of 60 to 80 percent, compared to a conventional electric power plant at 30 to 40 percent. A cogeneration plant's thermal efficiency (more accurately called an energy utilization factor by J.H. Horlock in his text Cogeneration—Combined Heat and Power) is generally calculated as the sum of the electric power output and the useful heat produced divided by the fuel energy supplied. But work output is more expensive and difficult to produce, while useful heat is easy to produce cheaply and has a lower thermodynamic availability.

A much better performance criterion for cogeneration plants is the fuel energy savings ratio, or FESR. It is defined as the ratio of fuel energy saved by use of the cogeneration plant to the fuel energy required to run the separate heating plant and power plant that the cogeneration facility replaces. The value of the ratio should be less than 1.0—or 100 percent, the unattainable ideal—and greater than zero.

Assuming natural gas as the fuel of choice for both heat and power, one can see that by purchasing electric power from a utility, a large institutional user such as a university or hospital is, in effect, burning two measures of fuel when it could burn one. That is, one measure of natural gas is combusted in an offsite utility company's gas turbines (and exhausted up a stack) and another measure of natural gas is burned in steam boilers (and exhausted up a stack).

Not your grandfather's boiler room: The lube oil assembly, the electric generator, the gas turbine enclosure, and the heat recovery steam generators (left to right).

An institutional cogeneration plant would use one measure of costly natural gas twice—first for generating electric power in a campus-installed gas turbine, and then passing the hot exhaust gases through a heat recovery system to produce steam (and only then exhausting up a stack). If surplus electric power were produced, the institution could easily sell it to the outside electric grid and make money to offset fuel costs.

And every penny counts these days. Total utility costs for universities such as UConn can be substantial. For fiscal year 2005, utility costs for electricity, natural gas, and fuel oil (which is used as a supplemental fuel source) amounted to $23.4 million for the Storrs campus. That's 20 percent of the campus operating budget and about 3 percent of the total campus budget. By investing in ways to cut the amount of useful heat being vented as waste, we could, in theory, produce real savings.

In 2002, the university put out a request for proposals for a completely new, from-the-ground-up, cogeneration plant, with the following requirements: that it produce a peak of 25 megawatts as well as 200,000 pounds per hour of steam and 6,000 refrigeration tons of chilled water. A contract was awarded and construction, which started in 2003, was completed in 2005 at a cost of about $80 million. The new UConn cogen plant went online earlier this year.

getting waste to work

The heart of the cogen plant consists of three 7.5-megawatt Solar Taurus 70 gas turbines, which have a rated thermal efficiency of 34 percent. Yes, a larger single gas turbine would have been more efficient—Pratt & Whitney's FT8, for instance, is rated at 25 MW and 38 percent thermal efficiency. But one larger turbine would lack the flexibility of three units to cover the wide load variations that a university experiences over the academic year. The load at UConn ranges from 8 to 22 MW.

The fuel of choice is natural gas, but the gas turbines can alternately be switched over to fuel oil. A major high-pressure natural gas line is located near the Storrs campus, so compression equipment, which might consume as much as 1 MW to inject gas into the combustors of the gas turbines, was not needed. The university also has 300,000 gallons of fuel oil as a backup in case of a natural gas supply disruption.

Each gas turbine drives a water-cooled electric generator. The exhaust, which runs as high as 900°F, passes into a heat recovery steam generator to produce both high-pressure and low-pressure steam. After passing through the HRSG heat exchangers, the gas turbines' exhaust, now at about 300°F, goes up a 120-foot stack to the atmosphere, after being treated with ammonia to reduce NO_x emissions. Each of the three HRSGs can also be fired with natural gas burners, if steam output needs to be increased.

Low-pressure steam is used for campus heating, distributed through utility tunnels and pipes to campus buildings, kitchens, and laboratories. During the warmer months when heating loads are greatly reduced, the low-pressure steam powers turbines that drive three refrigeration compressors that supply up to 6,000 refrigeration tons of chilled water to air conditioning units in campus buildings. In addition, a small percentage of the chilled water is used to reduce the gas turbine inlet air temperature during hot weather. Gas turbines are momentum changers, so outputs decrease as air temperatures rise and air density falls. Thus, on a 90°F day, gas turbine inlet air can be reduced to 50°F by chilled water heat exchangers, to maintain a near constant—and predictable—electrical output from each of the Solar gas turbines.

High-pressure steam from each HRSG is used to power one single-stage steam turbine that drives a 5 MW water-cooled electric generator, providing additional electric power and higher combined-cycle thermal efficiencies. The exhaust from the steam turbine, now at a reduced pressure and temperature, can be added to low-pressure steam to either heat or cool students and faculty. Thus, this portion of the energy conversion cycle in UConn's cogen plant has made three uses of a unit of gas turbine fuel.

The steam turbine exhaust can also be directed to a dump condenser if not needed for campus heating or air conditioning. The Second Law of Thermodynamics requires heat rejection for any power plant operating in a cycle, and that is accomplished by banks of water cooling towers mounted on the roof of the cogen plant.

The HRSG steam drum as seen from a third-floor catwalk. More and more universities are turning to combined heat and power systems.

The UConn cogeneration plant makes all kinds of sense, thermodynamically and environmentally. The new plant makes better and more efficient use of a high-grade fuel and has less of an environmental impact than the systems it replaces.

One can calculate a demand rate (ratio of heat required to electrical power required) using the plant design requirements—25 MW of electricity and 200,000 pounds per hour of steam. The equation for fuel energy savings ratio is given in Sir John Horlock's cogeneration text. That calculation shows that the FESR could be as high as 48 percent. Although this represents a rather extreme condition (maximum electrical and heating demands occurring at the same time), it shows that the cogen plant would use only 52 percent of the fuel consumed by conventional, non-cogeneration means.

Theories aside, will the UConn Cogeneration Plant actually save the university money? Due to a quirk in state law, the plant currently is not permitted to sell electric power to the outside grid. This may change in the future, when lawmakers become more enlightened, or as Connecticut's energy costs increase. A financial study done by consultants during the plant's planning phase shows definite savings over the long run, especially since the cost of electricity can be expected to vary with the cost of natural gas in New England. In the short term, the university system is still backed up by the outside electrical grid, and if the "spark spread" becomes unfavorable, electrical power can be purchased from grid sources.

To date, the plant has been online less than a year, with operations being interrupted by corrections of construction and assembly problems that occur with any new facility as complicated as this state-of-the-art cogeneration plant. Universities such as Rice, Stanford, MIT, and Florida have cogeneration plants that have proven their worth, once they were up and running.

a teachable moment

There is also a component of engineering education to the facility. The cogen plant is located right on campus. Students and faculty wandering by the plant are, by and large, unaware that it is in operation, even at full load, since it is quiet and has no visible stack exhaust. And, as the plant was being designed, it was advocated that it should have a classroom, for use by engineering classes and others to learn about energy conversion, firsthand. When approval for construction of the plant was sought from the university's board of trustees, this educational aspect proved to be a good selling point.

The classroom, located on ground level below the control room, has separate entrances for students and a number of PCs to monitor—but not control—cogen plant operations. Students can also observe overall plant operations (updated every two minutes) at a Web site, http://137.99.254.89/pe/cogenhome.htm.

Although there's not a little pride in the fact that we finally have our cogeneration plant here on campus, I can't help but feel that this is something we could have achieved a dozen years earlier. As the larger society faces the prospect of having to use fossil fuels more wisely in order to conserve a finite resource and to reduce the production of greenhouse gases, I can't help but worry that other large but sensible steps toward efficiency will also take as long.

Lee Langston is professor emeritus in the Department of Mechanical Engineering at the University of Connecticut in Storrs, and former editor of ASME's Journal of Engineering for Gas Turbines and Power.