TECHNICAL FEATURE

Dehumidifying
Natatoriums in Hot,
Extreme Humid Climates
BY SHOAIB AJMAL KHAN

Dehumidification is a highly energy-intensive process, particularly in hot and

humid climate zones. A space requiring strict humidity setpoints further introduces
complexity in devising an optimum dehumidification system. The most common
principle used for dehumidification is subcooling air using chilled water or direct-
expansion (DX) cooling coils. Dehumidification using desiccant-based systems is also
being touted as having many benefits. These statements alone may be valid when
there is no requirement for high ventilation rates and recirculation air changes to
maintain acceptable indoor air quality parameters. But most of the high latent load
spaces using such systems typically require high ventilation rates and recirculation
air changes. One of the most common applications of these systems is natatoriums,
the topic of this discussion.

To evaluate dehumidification systems for hot with moisture content reaching as high as 25.7 g/kg con-
and humid Climate Zones 0, 1 and 2 in Table A-3 of sidering 0.4% dehumidification dew point/mean coin-
ANSI/ASHRAE Standard 169-2013, Climatic Data for cident dry bulb (DP/MCDB) as per climate data from the
Building Design Standards, the United Arab Emirates 2017 ASHRAE Handbook–Fundamentals for Bateen station
was selected as the ideal weather criteria for hot and (World Meteorological Organization [WMO] ID: 412160).
extreme humid climate zones discussed in this study. This article uses Al Ain International airport data with
The discussion and results can be applied to hot and moisture content reaching 22.2 g/kg considering 0.4%
humid conditions around the world. dehumidification DP/MCDB (WMO ID: 412180). It estab-
The United Arab Emirates, classified as Climate Zone lishes four engineered cooling and dehumidification
0B per ASHRAE Standard 169-2013, has one of the high- system configurations, based on two different principles
est wet-bulb temperatures (30°C [86°F]) in the world, for dehumidification using multiple approaches and

Shoaib Ajmal Khan is a mechanical engineer at Obermeyer Middleast GMBH, Abu Dhabi, United Arab Emirates.

26 ASHRAE JOURNAL ashrae.org O CTO B E R 2020

 TECHNICAL FEATURE

strategies to optimize initial costs, energy costs and reli- As discussed earlier, the most common principle used
able control methodology. for dehumidification is subcooling air to the dew point,
These system configurations have been evaluated for inducing condensation of moisture content in the air
a 1500 m² (16,146 ft2) swimming pool hall (natatorium) using chilled water or DX cooling coils. However, there
with ceiling height of about 8 m (26 ft). The swimming can be a number of amalgamations and approaches to
pool hall consists of a 25 m × 20 m (82 ft × 66 ft) semi- this principle. The same applies to desiccant dehumidi-
Olympic-sized swimming pool with an evaporation load fication systems.
of about 105.6 kg/h (233 lb/h) or about 70 kW of latent For this analysis, four different system configurations
load due to evaporation (considering latent heat of evap- have been considered. The intent is to use the most typi-
oration at 2400 kJ/kg [1,031 Btu/lb]). This evaporation cal solutions offered in the HVAC market, while simul-
load is calculated using the empirical formula provided taneously achieving energy conservation by reducing
in the 2019 ASHRAE Handbook–HVAC Applications, Chapter energy on cooling and reheating more air volume than
6, “Indoor Swimming Pools.” is necessary. Mixing sections have been used in differ-
Parameters considered in this calculation are as fol- ent configurations to achieve the latter. These system
lows: activity factor 1; pool surface water temperature: configurations include the ventilation load being met by
26°C (79°F); air temperature over the pool water: 28°C the same dehumidification unit and considers energy
(82°F); air humidity over the pool water: 50%. The space recovery using enthalpy wheels as per Standard 90.1-
is maintained at 28°C (82°F) and 50% relative humidity 2016. The cooling and dehumidification system options
(RH). The pool has been considered a no-spectator zone included are briefly discussed here:
and, therefore, a minimum of 4 ach is the air delivery Option 1. Enthalpy wheel for energy recovery; chilled
rate per the 2019 ASHRAE Handbook–HVAC Applications, water cooling coil and dehumidification unit with elec-
Chapter 6. The pool and deck area ventilation rate is tric resistance heater.
2.4 L/s·m² (0.06 gallon/s·ft2) based on ASHRAE Standard Option 2. Enthalpy wheel for energy recovery; dedi-
62.1-2016. cated DX dehumidification unit with hot-gas reheat con-
These rates have been used during cooling load calcu- denser; a separate unit with chilled water cooling coil for
lations to define the supply air temperature for different sensible cooling.
system configurations, considering the supply tempera- Option 3. Enthalpy wheel for energy recovery; desic-
ture is always above the space dew-point temperature of cant dehumidification system using hygroscopic wheel;
16.6°C (61.9°F) (corresponding to a 28°C [82°F] dry-bulb and a post-cooling coil for sensible cooling.
setpoint and 50% RH). Option 4. Enthalpy wheel for energy recovery; pre-
Building envelope parameters are better than the cooling coil for sensible and latent cooling; desiccant
minimum required by ASHRAE/IES Standard 90.1-2016. dehumidification system using hygroscopic wheel and a
Overhead lighting is 8 W/m² (0.7 W/ft2). Screen and mis- post-cooling coil for sensible cooling.
cellaneous extra-low voltage equipment heat dissipa- Figure 1 shows system flow diagrams for each of these
tion is at 2000 W. Peak occupancy is assumed to be 150, options.
with the activity level of athletics. The cooling system All options consider the chilled water cooling coil is
is zoned as perimeter and internal zones, per Standard being supplied from an air-cooled chiller with an IPLV of
90.1-2016. 4.3. The total enthalpy wheel should be replaced with a

Author’s Note: This article is strictly based on an extremely hot and humid climate and utility costs associated with a specific geographical
location. As such, the results shall be read with the following cautions:
• Certain geographic locations may vary the stated results from a moderate to a much greater amount due to factors not discussed in this
article. These include but are not limited to the fact that some hot and humid climates may have freezing conditions during the year and
may need energy for freeze protection of some components, and added preheating of outdoor air during freezing winter conditions may
exist.
• Local code requirements that are different than what is covered/considered in the article, such as economizer requirements.
• Local utility costs may be higher than considered in the article in certain localities.
• Procurement costs for some systems could be higher or lower in certain locations due to local or international sourcing, affecting capi-
tal cost, which may impact life-cycle cost.

O CTO B E R 2020 ashrae.org ASHRAE JOURNAL 27

TECHNICAL FEATURE

sensible heat wheel in climate zones FIGURE 1 System flow diagrams for the four options discussed.
in which the outdoor air moisture
content or humidity is less than Space
Temp.: 28°C, ±2°C DBT
the indoor air moisture content or Exhaust Air RH: 50%, ±5%
humidity. Option 1: CHW Coil and 3100 L/s
AH: 11.4 g/kg
To put the highly energy-intensive Electric Resistance Reheat 28°C DB
11.4 g/kg
latent cooling process into perspec- Return Air
tive before discussing it further, the
sensible heat of air is 1.006 kJ/kg Enthalpy Wheel
(0.432 Btu/lb), while the latent 3400 L/s 6645 L/s
Exhaust 28°C DB 28°C DB
heat of vaporization is 2454 kJ/kg Air 11.4 g/kg 11.4 g/kg
(1055 Btu/lb). Exhaust Air
6500 L/s 6500 L/s 13 145 L/s
Fan
Option 1 is considered the baseline 28.7°C DB 11.5°C DB 25°C DB
12.5 g/kg 8.4 g/kg 9.9 g/kg
for all other options and is based Fresh Air
Intake 3255 L/s 3100 L/s 13 145 L/s
on the most common application 33°C DB 29.5°C DB 19.8°C DB
of the subcooling principle using 22.2 g/kg 13.8 g/kg CHW Cooling 9.9 g/kg Electric Resistance
Coil Heater
chilled water from building chill-
ers. Since the dehumidification Space
8320 L/s
21.5°C DB
process is dependent on cooling the Temp.: 28°C, ±2°C DBT 11.4 g/kg
Exhaust Air RH: 50%, ±5%
air to the dew point for moisture AH: 11.4 g/kg
Option 2: Decoupled DX
condensation, reheat is required to Coil for Latent Load with 8320 L/s Cooling
adequately maintain comfort condi- Hot-Gas Bypass and CHW
Coil for Sensible Load
3400 L/s 28°C DB CHW Coil
28°C DB 11.4 g/kg
tions at part-load conditions. The 11.4 g/kg
Return Air
reheat typically involves using an
electric resistance heater, which is a Enthalpy Wheel
violation of Clause 6.3.2 of Standard Exhaust
Compressor To/From
CV Backup
90.1-2016 and is a waste of energy on Air Condenser
3100 L/s
top of the energy-intensive subcool- Exhaust Air 28°C DB TMV 6500 L/s
Fan 28°C DB
ing process for dehumidification. 11.4 g/kg
8.4 g/kg
Fresh Air
In this option, multiple strategies Intake 3255 L/s 3100 L/s 6500 L/s 6500 L/s
have been used to reduce the source 33°C DB 29.5°C DB 28.7°C DB 11.5°C DB
22.2 g/kg 13.8 g/kg 12.5 g/kg Cooling 8.4 g/kg Heating
of energy consumption. The pri- DX Coil
Hot-Gas
Bypass
mary one is only treating the volume
of air required to meet the latent
load of the space corresponding to latent cooling can be optimized to a lower condensing
off-coil moisture content of 8.4 g/kg (1.2 gr/lb) at 11.5°C temperature compared to chillers. (Doing the same in
(52.7°F) dry bulb. Mixing sections have been used to mix Option 1 would increase the chiller lift and in turn lower
return air with the cooled supply airstream and raise the the efficiency of the whole building cooling system.)
temperature of air before passing through the electric Second, decoupling latent and sensible loads can save
resistance heater for sensible heat. energy at part-load conditions and give better tempera-
Option 2 uses a decoupled approach toward sensible ture and humidity control. Also, a more efficient chiller
and latent cooling. It uses a DX cooling coil for latent system is used for space cooling.
load and a more energy-efficient chilled water sensible Third, a hot-gas reheat condenser strategy has been
cooling process. This option has a three-fold advantage used to reheat the air after the dehumidification process
compared to baseline Option 1. First, the off-coil mois- in which the hot refrigerant gas leaving the compressor
ture content in a condensing coil depends on the cooling is channeled to a reheat condenser to heat the dehu-
coil dew-point temperature. A dedicated DX system for midified air. This reheat is free energy and is, in essence,

28 ASHRAE JOURNAL ashrae.org O CTO B E R 2020

 TECHNICAL FEATURE

instead of a constant/continuous
13 145 L/s
type compressor to closely match
25°C DB the load profile and precise controls
9.7 g/kg
Space in conjunction with hot-gas reheat
Temp.: 28°C, ±2°C DBT condenser modulation to consistently
Exhaust Air RH: 50%, ±5%
Option 3: Desiccant System AH: 11.4 g/kg Post-Cool maintain humidity levels as required
3100 L/s CHW Coil 13 145 L/s
for Latent Load and
28°C DB 36.2°C DB for the dehumidification process.
CHW Coil for Sensible
Cooling 11.4 g/kg Return Air 9.7 g/kg Another downside to this option can
Hygroscopic 7645 L/s be the less efficient DX system com-
Wheel 28°C DB
Enthalpy Wheel 11.4 g/kg pared to the building chiller used

Process Side Reactivation

Reactivation Fan
for latent load; this will be evaluated

Side
Exhaust Reactivation Air
Air 1815 L/s Inlet further in coordination with free
2400 L/s
Exhaust 28°C DB Reactivation energy provided by the hot-gas reheat
Air Fan 11.4 g/kg Heater
condenser.
Fresh Air
Intake 3255 L/s 3100 L/s 5500 L/s 5500 L/s Option 3 and Option 4 use a desic-
33°C DB 29.5°C DB 28.8°C DB 47.6°C DB cant dehumidification system based
22.2 g/kg 13.8 g/kg 12.8 g/kg 7.3 g/kg
on a hygroscopic wheel using silica
13 145 L/s gel as the desiccant media. It removes
25°C DB moisture content in the air by adsorp-
9.7 g/kg
Space tion (moisture is taken up by the
Temp.: 28°C, ±2°C DBT
Exhaust Air RH: 50%, ±5% surface of its numerous pores rather
Option 4: Hybrid Desiccant Post-Cool
System and Precool CHW Coil for 3100 L/s AH: 11.4 g/kg CHW Coil 13 145 L/s than being taken up by volume in the
28°C DB 27.6°C DB
Latent Load and Post-Cool CHW
11.4 g/kg Return Air 9.7 g/kg case of absorption). This desiccant
Coil for Sensible Load
Hygroscopic 10 045 L/s media typically has the capability to
Enthalpy Wheel
Wheel 28°C DB hold moisture quantity many times
11.4 g/kg
Process Side Reactivation

Reactivation Fan
its actual weight. Since this principle
Side

Exhaust Reactivation Air

Air 715 L/s Inlet of dehumidification does not depend
Exhaust
Air Fan Reactivation on dew-point temperature for off-
2100 L/s
3100 L/s 3100 L/s 12.8°C DB
Heater system moisture content, it can be
Fresh Air
Intake 3255 L/s 29.5°C DB 12.8°C DB 8.8 g/kg 2100 L/s 3100 L/s crucial for conditions requiring lower
13.8 g/kg 8.8 g/kg 33°C DB 26.5°C DB
33°C DB 1000 L/s 3.3 g/kg 5.1 g/kg
humidity levels and can result in
22.2 g/kg Precool CHW
Coil
12.8°C DB reducing the volume of air treated to
8.8 g/kg Internal Bypass
meet required latent loads.
However, although the volume of
complementing the DX dehumidification process by treated air can be reduced by using
giving the heat of rejection (latent energy gained from this principle, the total volume of air in the system
the evaporator plus the compressor heat) back to the air- would still be defined by required ventilation rates and
stream. A backup condenser is also used to control the air changes necessary to maintain acceptable indoor air
amount of heat rejected to dehumidified air; excess heat quality as per referenced standards. For this purpose, a
is rejected to atmosphere using a backup condenser. hygroscopic wheel is used in Option 3 and Option 4 in
The first downside to this system is due to the type of different system configurations.
compressors usually used in DX systems. The constant Option 3 is a basic desiccant-based system configura-
nature of these compressors adversely affects part-load tion in which a hygroscopic wheel reduces the moisture
dehumidification and gives sort of “gray regions” or content of process air after energy is recovered using the
“dead bands” in the dehumidification process during enthalpy wheel. The hygroscopic wheel is used solely to
compressor off-cycles. However, this can be mitigated process the air required to meet the latent load corre-
by using a modulating compressor with many steps sponding to a reduction of moisture content by 5.5 g/kg

O CTO B E R 2020 ashrae.org ASHRAE JOURNAL 29

TECHNICAL FEATURE

(0.79 gr/lb). For this purpose, a mixing section is

TABLE 1 Energy consumption cost comparison (kWh/yr).
used to pass on the required quantity of process
OPTION 1 OPTION 2 OPTION 3 OPTION 4
air to the hygroscopic wheel considering mois-
Supply Fan 225,137 225,115 225,137 225,137
ture content of about 7.3 g/kg (1 gr/lb) off the
Extract Fan 40,231 40,231 40,231 40,231
hygroscopic wheel.
The downside of this system is the need for Energy Wheel Motor 2,628 2,628 2,628 2,628

high grade heated regenerative air (typically CHW Coil 408,433 141,914 360,928 278,648
about 30% to 35% of process air at 100°C [180°F] DX Coil 627,236
above ambient) to regenerate the desiccant Electric Heater 769,022 18,887
media (silica gel) for the adsorption process, Reactivation Fan 8,833 3,475
which typically involves using an electric Hygroscopic Wheel Motor 2,628 2,628
resistance heater. Also, the processed air out Reactivation Heater 1,770,958 696,674
of the hygroscopic wheel is dry but heated to a Total Energy Consumption per Year (kWh) 1,445,452 1,056,012 2,411,343 1,249,422
higher temperature because of heat exchange Total Operational Cost ($/Year) $82,391 $60,193 $137,447 $71,217
with regenerative air. Regeneration is the most All device consumption measured in kWh/year. Rate = $0.057/kWh
energy-intensive process in this option on top of
a chilled water cooling coil (post-cooling coil) FIGURE 2 Energy consumption cost comparison.
used to sensibly cool the heated process air. 2.5
Option 4 is devised to minimize the energy 1,770,958
consumption from the regeneration process of 2
the desiccant system by using a hybrid chilled
water cooling coil and hygroscopic wheel for
Dollars (Millions)

1.5
dehumidification. The strategy is to further 769,022
dry the air from 7.3 g/kg (1 gr/lb) in Option 3 to 1
about 3.3 g/kg (0.5 gr/lb) by using a precooling 627,236
696,674
chilled water coil between the enthalpy wheel 0.5 408,433 360,928 278,648
and the hygroscopic wheel. The precooling 40,231
141,914
40,231 40,231
40,231
coil after the enthalpy wheel cools the air to 225,137 225,115 225,137 225,137
0
the dew-point temperature, condensing the Option 1 Option 2 Option 3 Option 4
moisture content in the air and giving drier air Note: All device consumption measured in kWh/year. Some energy consumers may not be visible due to
insignificant consumption or may not be applicable to a certain option.
at 8.8 g/kg (1.3 gr/lb) to the hygroscopic wheel,
Reactivation Heater Hygroscopic Wheel Motor Reactivation Fan Electric Heater
which further reduces the moisture content
DX Coil CHW Coil Energy Wheel Motor Extract Fan Supply Fan
to 3.3 g/kg (0.5 gr/lb). Note that since the air
became drier, this configuration needs less
volume of air to be treated to meet the latent load. This than necessary. This essentially reduces waste energy
reduces the overall size of the hygroscopic wheel sys- in the system. This will be further evaluated in the next
tem, which in turn reduces the energy required for the section to analyze the reduction in energy consumption
regeneration process. compared to the baseline.
The post-cooling coil will also require less energy
compared to Option 3 since the quantity of hot and dry Energy Consumption Comparison
process air mixing with return air has also been reduced To evaluate these options from an energy consump-
significantly. Since the outdoor fresh air intake for tion perspective, the bin method was used by computing
ventilation is defined by Standard 62.1-2016, the same bin hours based on ASHRAE International Weather for
quantity of air is treated at the precooling coil (after Energy Consumption. For the purpose of this evalua-
energy recovery from the enthalpy wheel) and then tion, a spreadsheet tool was developed by the author to
bypassed from the hygroscopic wheel to avoid wasting compute energy consumption at each step of all system
energy by treating more air volume for the latent load configurations. A basic energy model based on bin hours

30 ASHRAE JOURNAL ashrae.org O CTO B E R 2020

TECHNICAL FEATURE

was conducted using an hourly

TABLE 2 Life-cycle cost comparison.
analysis program for the build-
SUPPLY
ing envelope cooling load and was AIRFLOW
REQUIRED COOLING TOTAL CAPITAL TOTAL ANNUAL ENERGY 20-YEAR LIFE-CYCLE
CAPACITY (KW) COST ($) CONSUMPTION COST ($) COST NPV ($)
transferred to the spreadsheet tool (L/S)

for detailed computation. Latent Option 1 13,675 226 295,762 82,391 997,201
load from pool evaporation was Option 2 13,675 93.1 352,550 60,193 865,004
considered separately and was Option 3 13,675 185 377,712 137,447 1,547,872
assumed to be constant because no Option 4 13,675 173 346,376 71,217 952,687
pool cover is used. The cost of elec- Rates: $0.057/kWh, discount rate of 10%, chilled water plant room $1,200/ton of refrigeration.
trical energy was considered to be a
flat rate of $0.057/kWh (AED 0.21/kWh). FIGURE 3 Cost comparison.

Table 1 and Figure 2 show that Option 2 (with the decou- 1.8
pled latent load being supplied by a DX dehumidifica- 1.6 Total Capital Cost ($) 1,547,872
tion coil and hot-gas reheat condenser and sensible load 20-Year Life-Cycle Cost NPV ($)
being met by a chilled water coil) is the lowest energy 1.4
consuming option, consuming about 73% of energy 1.2
compared to the base case (Option 1), even though the Dollars (Millions)
997,201
DX system is less efficient than the chiller system. This 1 952,687
865,004
is mostly because of the free energy using the hot-gas 0.8
reheat condenser strategy to reheat the dehumidified
air, especially at part-load conditions, replacing 53% of 0.6
the total energy consumption by the electric resistance 0.4 352,550 377,712 346,376
295,762
heater in the base case (Option 1) that was wasted to
reheat the dehumidified air. 0.2
Option 3 (using a basic desiccant dehumidification and 0
cooling system) is the most energy consuming configu- Option 1 Option 2 Option 3 Option 4
ration of the evaluated options, with about 66% more
energy consumed compared to the base case (Option 1). because of the reduced volume of air (55% compared
This is chiefly because of the massive amount of energy to Option 3) that needs to be treated to meet the space
used to regenerate desiccant media for the adsorption latent load.
process using an electric resistance reactivation heater.
This draws about 73% of the total energy consumed in Life-Cycle Cost Comparison
this option. The life-cycle cost comparison considers the capital
The next in line after Option 2 in terms of reduced cost as well as operating costs or annual energy con-
energy consumption, compared to the base case, is sumption costs. Capital cost includes equipment costs
Option 4 (using a hybrid of a chilled water coil and and associated chilled water plant costs. Equipment
desiccant dehumidification system). About 86% of the costs have been based on selections and quotations
energy is consumed compared to the base case. This received from various major manufacturers, while the
shows that a minor modification to Option 3 (add- chilled water plant cost is based on $1,200/ton of refrig-
ing a precooling coil to reduce the volume of air to be eration. A 20-year life-cycle cost is computed based on
treated) pays off in reducing the total energy consumed the net present value (NPV) of annual energy consump-
by the reactivation heater. What is interesting, though, tion cost assuming a 10% discount rate.
is that the total energy consumed by chilled water The results shown in Table 2 and Figure 3 mimic the
cooling coils (precool for latent and sensible cooling, results discussed in the energy consumption cost com-
post-cool for sensible cooling) in Option 4 is less than parison and show that Option 2 results in the lowest
the energy consumed by the only chilled water cooling life-cycle cost compared to the base case (about 13%
coil (post-cool for sensible cooling) in Option 3. This is lower). Option 3 has the highest life-cycle cost (about

32 ASHRAE JOURNAL ashrae.org O CTO B E R 2020

 TECHNICAL FEATURE

55% higher). Option 4 is much the same as the base case, use an electric heater for regeneration result in the
with about 95% of the life-cycle cost of the base case. highest life-cycle costs. However, the hybrid desiccant
In terms of capital costs, Option 3 (basic desiccant sys- and chilled water cooling coil option will be the lowest
tem) has the highest cost followed by Option 2 (decou- energy consumption configuration, provided a sustain-
pled DX dehumidification and chilled water cooling) able source such as solar heating or other means of free
followed by Option 4 (hybrid desiccant and chilled water energy could be used for regenerative heaters. This can
dehumidification). The lowest capital cost is the base be derived from Figure 2.
case Option 1 (chilled water cooling and electric resis-
tance reheat). These results, when read in conjunction References
with life-cycle costs, show that although the capital cost 1. Ribeiro, E., H.M. Jorge, D.A. Quintela. 2011. “HVAC system
energy optimization in indoor swimming pools.”Proceedings of the
of evaluated options may be greater than the base case, 2011 3rd International Youth Conference on Energetics (IYCE) 1 – 7.
life-cycle cost is hugely dependent on energy consump- 2. Pillai, J., R. Desai, A. Ten. 2018. “Dehumidification strategies
tion costs because of the energy-intensive dehumidifica- and their applicability based on climate and building typology.”
2018 Building Performance Analysis Conference and SimBuild 759 – 766.
tion process. 3. Lochner, G., L. Wasner. 2017. “Ventilation requirements for
indoor pools.” ASHRAE Journal (7).
Conclusions 4. ANSI/ASHRAE Standard 90.1-2016, Energy Standard for Buildings
Except Low-Rise Residential Buildings.
Based on the results and discussions presented in 5. ANSI/ASHRAE Standard 62.1-2016,
this article, it can be concluded that dehumidification Ventilation for Acceptable Indoor Air Quality.
systems cannot be analyzed from a straightforward 6. ANSI/ASHRAE Standard 169-2013, Climatic
Data for Building Design Standards.
approach on the working principle of different systems. 7. 2019 ASHRAE Handbook–HVAC Applications. https://bit.ly/3bajige
Certain principles or strategies might reduce the sys- 8. 2017 ASHRAE Handbook–Fundamentals. Rate this Column

tem size by reducing the volume of air being treated to

meet latent loads, but a more extensive analysis needs
to be undertaken because of other parameters such as
required ventilation rates, recommended air changes
per hour for acceptable indoor air quality and higher
supply air temperatures to avoid condensation. These
determine the energy consumption and, in turn, the
life-cycle cost of dehumidification systems.
Additionally, it is fundamental to ensure that the
dehumidification system is capable of consistently
maintaining the required humidity levels at all load
conditions.
The following conclusions can be drawn for dehumidi-
fication system options for natatoriums in the hot and
humid climate zones presented in this discussion:
• Decoupled DX dehumidification and chilled water
cooling with hot-gas reheat condenser (Option 2) pro-
vides the lowest energy and life-cycle costs. This con-
figuration can also provide efficient temperature and
humidity controls, with a humidity sensor controlling
the DX system and a temperature sensor controlling the
chilled water cooling coil. However, the DX system must
use a carefully selected modulating compressor with
precise controls to consistently follow the latent load
profile with the hot-gas reheat condenser strategy.
• Desiccant-based dehumidification systems that

O CTO B E R 2020 ashrae.org ASHRAE JOURNAL 33