Restroom Exhaust Systems: Column Engineer'S Notebook
Restroom Exhaust Systems: Column Engineer'S Notebook
Restroom Exhaust Systems: Column Engineer'S Notebook
This article was published in ASHRAE Journal, February 2014. Copyright 2014 ASHRAE. Posted at www.ashrae.org. This article may
not be copied and/or distributed electronically or in paper form without permission of ASHRAE. For more information about ASHRAE
Journal, visit www.ashrae.org. Steven T. Taylor
Designing restroom exhaust and makeup air systems should be a trivial matter, but
many designs are expensive and inefficient. This month’s column outlines some tips
to design a system that performs effectively yet has low first costs and energy costs.
Figure 1 is an example of a poor design of a restroom Furthermore, ACH is inconvenient from a design
exhaust and makeup air system for a typical nonresi- perspective since rates must change as the design pro-
dential men’s restroom. The design is summarized as gresses if the size of the room or ceiling heights change.
follows: A more rational metric for determining ventilation
• The exhaust rate is based on 10 air changes per hour; rates in restrooms is airflow per plumbing fixture (e.g.,
• Air is exhausted above each water closet and urinal urinal or water closet) as prescribed by ASHRAE Standard
bank; and 62.11 and the model building codes that are based on
• Makeup air is sized at 90% of the exhaust rate and Standard 62.1 requirements. Standard 62.1-2013 requires
supplied by a constant volume reheat box connected to an exhaust rate of 70 cfm (35 L/s) per water closet/urinal
the building’s cooling system. where long periods of heavy use are expected to occur,
A much less expensive and more energy-efficient e.g., restrooms in theatres, schools, and sports facilities,
design is shown in Figure 2. Its design characteristics are and 50 cfm (25 L/s) in other applications, such as office
discussed below. buildings. For simplicity, and to be conservative, the
higher rate is often used for all applications.
Exhaust Rate Sizing In our example restroom, the exhaust rate at 70 cfm (35
In the example design in Figure 1, the exhaust rate was L/s) per plumbing fixture would be 420 cfm (210 L/s) (Figure
based on air changes per hour (ACH), a rate propor- 2), down from 560 cfm (280 L/s) based on 10 ACH (Figure 1).
tional to the volume of the restroom. Using ACH to size
any ventilation system is fundamentally flawed:* Number of Exhaust Grilles
• It implies that the rate of air needed to dilute odors The design in Figure 1 includes four exhaust grilles, one
in the room is proportional to the volume of the room. over each water closet and one over the bank of urinals.
In fact, the ventilation rate is proportional to the source It may seem to make sense to exhaust above each fixture
strength of odors which has no relationship to the vol- since that is where odors are emitted, but it actually does
ume of the room. Would there be more odors to dilute not improve ventilation effectiveness because the ceiling
if the entry vestibule were larger or if the ceiling were grilles cannot capture the odors before they diffuse into
higher? Clearly not. the space.
• It requires larger ventilation rates for rooms with Figure 3 shows a computational fluid dynamics simula-
larger volumes. Under steady-state conditions, the tion of a typical exhaust grille. Note the velocity vectors
required ventilation rate is the same for two rooms with are only high near the grille; 2 ft or 3 ft (0.6 m or 0.9 m)
identical source strength regardless of volume. Under away from the grille face, the velocity vectors are zero
non-steady-state conditions, the ventilation rate is which means that odors produced closer to the floor
smaller, not larger, for rooms with larger volumes since level will not be captured by the grille.** Thus, there is no
pollutants have more volume in which to mix. value to placing an exhaust grille over each toilet; using
*This is true of hospitals and laboratories as well, but ACH has been used for so long there is
bureaucratic momentum to retain it. Airflow rates per unit area or some other metric that comes Steven T. Taylor, P.E., is a principal of Taylor Engineering in Alameda, Calif. He is a mem-
closer to scaling with the source strength of contaminants should be used. ber of SSPC 90.1 and chair of TC 4.3, Ventilation Requirements and Infiltration.
** This behavior is described in the old expression, “You cannot suck out a match,” or at least not without burning your lips.
Noise transfer may also be a concern if the adjacent space FIGURE 4 “Plaster frame” for lay-in grille used for access to balancing damper.
is anything but a corridor or other non-noise-sensitive
Lay-In Diffuser
space.
All makeup air is from the adjacent space so the rest-
room will be warmer based on the exhaust rate and the
lighting load (occupant loads are sufficiently transient, Plaster Frame
so they can generally be ignored). If only warmer by 1°F
or 2°F (0.5°C to 1°C), temperatures are generally con- must be coordinated with the architect to be sure they are
sidered acceptable and are often preferred particularly included in door schedules on architectural drawings.
at water closets. In the past, lighting loads in restrooms 2. Ducted transfer to the ceiling return air plenum (Fig-
were high enough that this approach resulted in overly ure 5b). This option will result in even warmer restroom
warm restroom temperatures. But energy codes have temperatures than the previous option if the return air
reduced lighting power to 1 W/ft2 (0.1 W/m2) (Standard plenum temperature is above space temperatures due
90.1) and 0.6 W/ft2 (0.06 W/m2) (California’s Title 24 to the heat from recessed light fixtures. However, the re-
Energy Standards3). At 0.6 W/ft2 and at the exhaust rate duced lighting power and the increasing use of pendent
shown in Figure 2, the temperature rise is 1.6°F (1°C) light fixtures in many modern buildings results in little
above adjacent spaces, which is likely to be acceptable. temperature rise, making this option viable.
The undercut or door louver must be sized so that the Sizing the transfer grille is somewhat tricky because
differential pressure across the restroom door is not so the return air plenum is also negatively pressurized. If
high that it will cause whistling or keep the door ajar. A it is too negative relative to the adjacent space (close to
pressure differential of 0.08 in.w.g.
(20 Pa) is generally acceptable. This
equation can be used to determine
approximate transfer-free area:
Q C
0.5
A= (1)
4005 DP
the 0.08 in.w.g. [20 Pa] maximum FIGURE 5 Other transfer air options.
across the door), transfer from the
return air plenum is not possible. If
the plenum pressure is 0.05 in.w.g. A B
(12 Pa), the pressure drop through
the transfer grille and duct, includ-
ing entry losses into the duct, must
be less than 0.03 in.w.g. (7.5 Pa) to
keep the total differential pressure
below 0.08 in.w.g. (20 Pa). With
perforated face grilles, this can be
achieved with a duct velocity of less
than about 325 fpm (1.65 m/s).
3. Ducted transfer to the adjacent
space (Figure 5a). This option results
FIGURE 6 Transfer air plus cooling supply options.
in the same space temperature as
Option 1. As with the undercut, the A B
grilles and ducts must be sized for a
total pressure drop of less than 0.08
in.w.g. (20 Pa). With perforated face
grilles, this can be achieved with a
duct velocity less than about 375 fpm
(1.9 m/s).
4. Ducted transfer to ceiling re-
turn air plenum with integrated
supply air (Figure 6a). When the
space will be too uncomfortable
without some cooling (or heating if the room is on the 5. Ducted transfer to ceiling return air plenum with
building perimeter), conditioned air must be supplied separate supply air (Figure 6b). If the design in Figure 6a is
along with transfer air. This typically can be air from felt to be too confusing, conditioned air may be supplied
an adjacent VAV zone; there is little benefit to provid- by a separate diffuser. This option is more expensive than
ing a separate zone for the restroom since most of the Option 4 but easier to understand and to balance.
air supplied to the room is neutral transfer air which
buffers any temperature excursions. The supply air Conclusions
rate should be sized just for the cooling or heating Traditional restroom exhaust and makeup air systems
load, 40 cfm (20 L/s) in this example, with the remain- can be both expensive and energy inefficient. The tips in
der of the makeup air supplied through the transfer this month’s column show how to reduce first costs and
opening. If the adjacent zone is VAV, the makeup air minimize energy use.
must be sized for the airflow required when the VAV
zone airflow is at its minimum. References
The design shown in Figure 6a is a bit unusual and often 1. ASHRAE Standard 62.1-2013, Ventilation for Acceptable Indoor Air Quality.
2. ASHRAE Standard 90.1-2013, Energy Standard for Buildings Except
generates a clarification request from mechanical con- Low-Rise Residential Buildings.
tractors: the conditioned supply air from the adjacent 3. 2013 Building Energy Efficiency Standards for Residential and
zone is ducted into a standard wye fitting with one end Nonresidential Buildings, Title 24, Part 6 CEC-400-2012-004-CMF.
4. “Discussion of the Use of Transfer Grilles to Facilitate Return Air Flow
open at the return air plenum. The negative pressure
in Central Return Systems.” http://tinyurl.com/o4sghcf; and “Discus-
in the restroom ensures that the supply air will not go sion of the Use of Transfer Grilles to Facilitate Return Air Flow in Cen-
backwards into the plenum. tral Return Systems.” http://tinyurl.com/ovabs66 are two examples.