Mark A. DeSorbo
SAN JOSE, CA — A cool $320,000 annual savings in operating costs may not seem like a big deal for the owner of a 64,000-square-foot cleanroom, but what he or she might find more interesting is all they have to do is decrease air velocity.
And not only is a savings shown, but the findings outlined in “The Study of Altering Air Velocities in Operational Cleanrooms” at CleanRooms West '99 in November also indicate that a decreased air velocity in the cleanroom lowers particle count.
“The industry philosophy says to turn up the air velocity to lower particle count, but I propose a different philosophy,” says Maribel Vazquez, a mechanical engineer who presented findings that evolved from research she performed at MIT in a Class 10 cleanroom while studying for her master's degree.
An 8 x 10 x 8-foot section of the cleanroom, she says, was isolated from the remainder of the facility for testing. The area encloses nine HEPA filters and one blank panel, as well as two wall-mounted return grilles and one floor grille. Plastic sheeting was placed along the open perimeter to simulate walls during two experiments.
The first experiment, Vazquez explains, measured velocity profiled within the designated clean space. Hot wire anemometers were used to measure fluctuating velocities from 50 to 70 to 90 feet per minute. Eight hot wires were mounted onto a probe rake and connected to a monitor. Anemometers were paired and placed at 1-, 3-, 5- and 7-feet elevations from the ground, with the seventh and eighth being on the top and just 8-inches from the HEPA ceiling. The first and second probes were on the bottom and about 7 feet from the ceiling.
She found that at each velocity, there was a stagnant air zone in the middle of the test area. The height of the zone increased as air velocity was decreased. That finding was critical because particles, Vazquez says, caught within the zone were not being exhausted.
“This isn't rocket science,” says the laughing Massachusetts Institute of Technology doctoral candidate, “but by reducing airflow velocity, you will have fewer particles near the product, but more at side wall exhausts. As long as your workers are walking slowly, like they should, and not doing calisthenics around the room, particles from the stagnant zone will not get on the product.”
The second experiment used three semiconductor wafers that were placed in triangular formation, 20 feet apart. Vazquez used a humidifier to blast 700,000 particles into the space. Air velocities were again altered from 50 to 70 to 90 feet per minute.
“I put the wafers on adjustable seats and raised them to 36 inches, tabletop level. The humidifier introduced particles for 90 seconds and then was turned off for another 90 to allow the particles to settle,” she says. “I was able to see the particles by using black drapes.”
As a result, Vazquez found once again that higher air velocities increased the number of particles in the test area. Lower velocities decreased the number at the work area level, but an increased number of particles were found at sidewall exhausts.
While her findings seem to defy industry practices, presentation attendees seemed receptive to the idea.
“There are tremendous advantages to this,” says John Petitt, facilities manager at Anadigics, a gallium arsenide fab in Warren, NJ. “It cuts energy consumption, it's quieter in the cleanroom and it extends the life of mechanical equipment,” he says.