Cleanrooms: Can They Be Energy-Efficient?
By Allan Chen, A_Chen@LBL.gov January 30, 1997

BERKELEY, CA -- Cleanrooms use tremendous amounts of energy, but until recently, there has been little effort to examine their energy use. Asking whether there was any way to make these facilities more energy-efficient, three investigators at the Ernest Orlando Lawrence Berkeley National Laboratory (Berkeley Lab) conducted a pilot project to test methods of reducing energy use.

Their study is part of an ongoing effort at Berkeley Lab to develop energy-efficient design and operational practices for cleanrooms.

"Cleanrooms are used extensively in high technology industries and research laboratories," says David Faulkner of Berkeley Lab's Energy & Environment Division. "These facilities maintain low particle concentrations by circulating the entire volume of air in the lab at a high rate, often 400 to 600 times per hour through high efficiency particulate air filters (HEPA)." The cost of energy to maintain the high recirculation rate, and to force the air through the filters adds up.

"In the typical cleanroom, air flows through the space at 60 to 100 feet per minute, and the fans are never turned off. Most clean rooms run 24 hours a day, seven days a week," says Faulkner.

The airflow keeps the number of particles in a volume of air below a certain minimum range. For example, a Class 100 cleanroom must have fewer than 100 particles of half a micron size or larger for each cubic foot of air. According to Faulkner, "common industry practice is to keep the fans on even when the cleanroom is not being used, because it is thought that if they're turned off, or if the fan speeds are changed, it will take hours to restore the room to the required particle specification."

To test this belief and determine whether an alternative airflow environment could maintain particle concentrations at nominal levels, Faulkner and William Fisk of the Indoor Environment Program, and John Walton of Berkeley Lab's Engineering Division developed a scheme called demand-control filtration (DCF).

Using a 300-square-foot cleanroom dedicated to experimental work, they set up a commercially available particle counter near a manufacturing station, and devised a system to feed the particle count back to an automatic control system that regulated the speed of the fan serving the room.

"As the measured particle concentration rose above a specified limit," says Faulkner, the fan speed increased automatically. When the concentration fell below the limit, the fan speed decreased." The fan motor consumes less energy when it runs at a lower speed, and the control system ensures that the fan is running only as much as is needed to keep the particle level in the air within the facility's specified limits.

The researchers first ran measurements of particle concentration under the facility's normal operating environment--fans running at high speed for 12 hours a day, and at a slightly lower speed during the night when the facility was not in use. Then they ran the fan under two other automatic control schemes.

In the first scheme, the counter measured particle concentration each second. At any time, if the count exceeded the maximum allowable concentration, the fan speed increased ten percent. If the count was below the allowable limit, the fan speed decreased by 0.1 percent. This scheme increased the fan speed quickly, and decreased it slowly.

In the second scheme, the fan speed increased in proportion to how many particles were in the air if the particle count was above the allowable threshold. Thus, if the counter detected substantially more than 100 particles per cubic foot, the control system increased fan speed by up to 70 percent; if the counter measured a slight increase in particles, the fan speed increased, but only up to about ten percent of its original speed.

"Two results stand out," says Faulkner. "The first is that both methods saved energy--60 percent of the fan energy used in the pre-existing control method, in which a variable-speed drive reduced fan speed at night. A more common situation is when the motors in a control scheme don't have variable speed drives. In this case, both of our methods would have saved 84 percent of the baseline energy. Both methods showed about the same energy savings."

Variable speed drives allow a motor to increase or decrease its speed smoothly. A motor with variable speed can use less energy than one without because it can use only as much energy as it needs to do its work.

"Second, when we increased and decreased the fan speed in the unoccupied cleanroom by ten percent or more, we did not see any sudden increases in particle concentration from the shedding of particles by the filters," he adds. This result suggests that managers of cleanroom facilities do not need to keep fans on their highest speed all the time to keep the air at the required level of purity.

Demand-controlled filtration may not work for cleanrooms that require the lowest possible level of airborne particles (Class 1), or production cleanrooms where operators want to maintain as clean an environment as possible. Further research may find the thresholds of acceptable cleanliness to which DCF could be optimized for a broader range of applications.

The results of this research appear in the November/December issue of the Journal of the Institute of Environmental Sciences. More information is available on the Web at http://eande.lbl.gov/CBS/NEWSLETTER/NL11/cleanrooms.html.

A report on California laboratory-type facilities including cleanrooms is also available at http://eande.lbl.gov/CBS/reports.html.

Berkeley Lab is a U.S. Department of Energy national laboratory located in Berkeley, California. It conducts unclassified research and is managed by the University of California.

Search | Home | Questions