ASHRAE Journal - June 2020 - 18

TECHNICAL FEATURE

840 cfm (396 L/s), which equates to 7.8 ach. The maximum exhaust flow rate with fume hood sashes open is
3,130 cfm (1477 L/s), equal to 29.1 ach.
In keeping with the applicable codes and standards,
the exhaust air from the labs is not recirculated. A central roof-mounted air handler provides 100% outdoor
air as makeup for the exhaust system. The AHU includes
MERV 7 prefilters, MERV 13 final filters, a hydronic heating coil, a hydronic cooling coil, a fan wall array of four
direct-drive supply fans with integral sound attenuators and variable frequency drive for each fan motor.
Multiple supply fans ensure continued system operation even with the failure of one fan. The system also
includes a heat pipe energy recovery coil.
System exhaust is provided using three roof-mounted,
induced flow/mixed flow fans, each sized at 50% of peak
design airflow. The multiple fans provide system redundancy as well as a high degree of turndown for periods
of low demand. The fans sit on an intake plenum that
includes fully modulating bypass dampers. The fan
motor speeds modulate to maintain system airflow via
duct static pressure, while the bypass dampers modulate
to maintain the fan discharge nozzle velocity at or above
3,000 ft/min (15 m/s). The high discharge velocity combined with the fan design itself results in an effective
exhaust plume height of more than 40 ft (12 m), ensuring no re-entrainment of the exhaust air into the intakes
of the air handler located on GPCL or adjacent buildings.
Supply and exhaust air in each lab are controlled using
venturi airflow valves. These are preferable to standard
VAV terminal units since they measure static pressure
very precisely, are highly calibrated at the factory and
have actuators able to respond to a change in demand in
less than 1 second. Each lab has a single supply air valve
that is coupled with both hydronic heating and cooling
coils, as well as individual exhaust valves for each fume
hood, general exhaust and snorkel exhaust-a total of
seven valves per lab.
Since the facility contains an office area as well as lab
and support space, maintaining pressure relationships
between the different types of occupancies is necessary
to ensure air from the labs does not migrate to other
areas. PNNL has used both pressure control and volumetric control in other facilities and found that because
pressure control is difficult to achieve and maintain,
volumetric control is generally preferable. At GPCL,
the use of highly accurate and reliable venturi valves
18

ASHRAE JOURNAL

ashrae.org

J U N E 2020

in conjunction with robust DDC controls ensures accurate control of supply and exhaust airflow necessary to
maintain a volumetric difference between supply and
exhaust. This difference ensures that the offices can be
maintained at a positive pressure with respect to both
the building exterior and adjacent labs.
The labs are likewise maintained at negative pressure
relative to the lab corridor. The volumetric offset in each
lab was set at approximately 15% of the minimum design
flow rate and is maintained by modulating the general
exhaust venturi valve position in response to the action
of the supply and fume hood exhaust valves as necessary to maintain the offset setpoint. The general exhaust
valve also accommodates periods when lab supply airflow exceeds exhaust, such as when supply for space
cooling exceeds the fume hood exhaust.
System supply airflow is controlled based on main duct
static pressure. The duct static pressure setpoint is reset
as necessary to provide makeup for the total lab exhaust
airflow, which varies according to the main exhaust
duct static pressure necessary to satisfy current airflow
at all fume hoods. The design was based on a total of 41
fume hoods with a diversity factor of 0.7, i.e., system fan
capacity was sized to accommodate 29 hoods at full flow
(sashes open) and 12 hoods at minimum flow.
Determining the appropriate diversity factor is not
easy and can vary widely between facilities. Fume hood
use at other PNNL facilities ranges from near zero to
essentially 100%. A factor that is too high will result in
primary equipment and ductwork that is oversized,
inefficient and unnecessarily costly, while a factor that
is too small will result in low flow at fume hoods and
potential loss of contaminant containment. At GPCL, the
0.7 factor was selected based on anticipated number of
researchers per lab, type of research, anticipated fume
hood use and comparison with other similar PNNL facilities. Useful resources on the subject include ASHRAE
publications2 and manufacturer's literature.3
Chilled water is supplied to GPCL through buried
piping from a central plant. The system supplies water
to the main air handler coils and terminal units at a
supply temperature of 42°F to 44°F (5.6°C to 6.7°C).
Since research labs frequently use water-cooled equipment, a separate process cooling loop was provided.
The process loop supplies water at 55°F (13°C) using a
brazed plate heat exchanger. Separating process from
HVAC avoids issues with condensation in the labs and

https://www.ashrae.org/

ASHRAE Journal - June 2020

Table of Contents for the Digital Edition of ASHRAE Journal - June 2020