By Mike Fitzpatrick and Ken Goldstein, Ph.D.

Mike FitzpatrickClick here to enlarge image

Much has been written recently regarding cleanroom airflow and airflow velocities. This interest is prompted by the desire to balance the contamination-control requirements of facilities with the energy consumption associated with moving the large quantities of air required.

Ken Goldstein, Ph.D.Click here to enlarge image

An article in the January 2003 edition of CleanRooms by Peter Rumsey makes the case that airflow velocities are frequently higher than they need be and recommends lowering these velocities where possible. While we certainly support this approach, the article prompts us to further muddy the waters by exploring some additional aspects of cleanroom airflow.

Most discussion regarding energy conservation and cleanroom airflow tends to focus on the average velocity of the air. It's here that the most significant savings in energy consumption can be realized. The velocity of the air is often determined by the degree of contamination control we wish to achieve—as a general rule, cleaner rooms require more air than rooms that are less clean. This approach looks primarily at the cleanliness aspects of the cleanroom without examining the other aspects of cleanroom air.

In addition to controlling airborne particles, cleanroom airflow is used to control room temperature, humidity and pressure. These parameters are usually controlled by the makeup air unit that introduces conditioned air into the recirculated air stream. IEST RP 12.1 Considerations in Cleanroom Design describes three common HVAC configurations for cleanrooms based upon the dominant requirement that determines the design of the HVAC system:

1) Exhaust-determined HVAC configurations
2) Cooling or heating-determined HVAC configurations
3) Cleanliness-determined HVAC configurations

Let's take a look at “cooling or heating-determined HVAC configurations.” These cleanroom HVAC systems are determined primarily by the cooling or heating requirements of the room. In nearly all cleanroom applications, the driving factor is the cooling requirement rather than the need for heat.

This system provides for the recirculation of a portion of the room air and the introduction of make-up air. Often the make-up air quantities used to replace exhaust air and to pressurize the room are insufficient to provide the necessary cooling.

Recall that the only acceptable way to deliberately introduce air into a cleanroom is through the filters. If the cooling air required exceeds the total amount of air that can be supplied through the filters, then additional filters must be used even though they are not required for airborne cleanliness. For this reason, we often see cleanrooms with velocities that seem higher than necessary considering only the cleanliness class of the room.

The classic example of this has long been the ISO Class 7 probe room used for the testing of integrated circuits. Glancing at the chart in Institute of Environmental Sciences and Technology (IEST; Rolling Meadows, Ill.) Recommended Practice (RP)-12, we would estimate that an average air change rate of 60 to 90 air changes per hour (AC/H) would be sufficient to achieve the ISO Class 7 conditions.

Equivalently, we could call for an average velocity of 0.05 to 0.08 meters/second (10 to 15 feet per minute) to achieve the same goal. But looking up at the ceilings of these rooms, we quickly see a ceiling coverage typically in the range of 20 to 40 percent that generate air change rates of up to 240 AC/H, or velocities of up to 0.2 m/s (40 fpm). Why the discrepancy between the air change rate that should be adequate for our cleanliness requirement and that actually used?

The answer lies in the nature of the equipment inside the room. Jammed full of electronic testers, these rooms require little exhaust. However, the testers generate a high heat load and the room's temperature must be controlled.

To achieve this a certain amount of “cold” air must be injected into the room. There is an upper limit to the amount of air we can push through each filter; and if we need more air for cooling, the only way to reasonably accomplish this is by adding more filters.

The result is that the heat load often determines the airflow velocity in a cleanroom, not the room's airborne cleanliness requirement. Cleanrooms with heavy heat loads often certify at an order of magnitude (10X) cleaner than the cleanliness specification requires. Rooms that are required to be ISO Class 7 often test at between ISO Class 5 and ISO Class 6.

This may seem like a waste of energy, but it occurs because of the temperature control requirements, not because someone “over-specified” the system. In terms of air cleanliness, we appear to be getting more than we wanted but don't blame it on the filters or the room's “excessive” airborne cleanliness. Instead, remember that “things” other than cleanliness may drive the design of the HVAC system.

Michael A. Fitzpatrick is program director of microelectronics for Lockwood Greene. A senior member of the IEST, he is chairman for WG012 (Considerations in Cleanroom Design) and WG028 (Minienvironments).