Available online at www.sciencedirect.

com

ScienceDirect
Availableonline
Available onlineatatwww.sciencedirect.com
www.sciencedirect.com
Energy Procedia 00 (2017) 000–000
ScienceDirect
ScienceDirect
www.elsevier.com/locate/procedia

Energy
EnergyProcedia 142
Procedia 00(2017)
(2017)1609–1614
000–000
www.elsevier.com/locate/procedia

9th International Conference on Applied Energy, ICAE2017, 21-24 August 2017, Cardiff, UK

Experimental analysis of a portable atmospheric water generator by

The 15th International Symposium on District Heating and Cooling
thermoelectric cooling method
Assessing
Shanshan theHe
Liua, Wei feasibility
b
*, DengyunofHuusing
a
, Song the
Lva, heat demand-outdoor
Delu Chen b
, Xin Wub, Fusuo
temperature function for a long-termXub,Sijia Libdistrict heat demand forecast
a
University of Science and Technology of China, No.96, Jinzhai Road, Hefei, 230026, China
I. Andrića,b,c*, A.
HefeiPinab
a
, P.of Ferrão
University
a
, J. Fournier
Technology,No.193,
b
Tunxi Road,.,Hefei,
B. 230009,
Lacarrière c
China , O. Le Corre
c

a
IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal
b
Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France
c
Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France
Abstract

A portable water generator (7 kg) with two thermoelectric coolers (TECs) was designed and experimentally investigated in this
study. Different inlet air relative humidity (RH) and air flow rates were investigated to obtain their impacts on the amount of
Abstract
generated water and condensation rate. The amount of generated water and condensation rate increased with the RH rose. The
amount
Districtofheating
generated water increased
networks with air
are commonly flow ratesinrose
addressed the but the condensation
literature rate most
as one of the had opposite
effectivetrend. The maximum
solutions amount
for decreasing the
2
ofgreenhouse gas emissions from the building sector. These systems require high investments which are returned through thesize
generated water was 25.1 g each hour with 0.216 m condensation surface and 58.2 W input power. This system had small heat
and could
sales. Duework at small
to the changedair climate
flow rates which was
conditions andsuitable
buildingforrenovation
outdoor use. This study
policies, had a guiding
heat demand in the role in could
future designing and
decrease,
optimizing a high-efficient portable water
prolonging the investment return period. generator.
The main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand
©forecast.
2017 The Authors.
The districtPublished by Elsevier
of Alvalade, locatedLtd.
in Lisbon (Portugal), was used as a case study. The district is consisted of 665
Peer-review under
buildings that responsibility
vary of the scientific
in both construction periodcommittee of theThree
and typology. 9th International Conference
weather scenarios (low,onmedium,
Appliedhigh)
Energy.
and three district
renovation scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were
Keywords: Portable atmospheric water generator; thermoelectric cooling; atmospheric vapor condensation; water generation rate
compared with results from a dynamic heat demand model, previously developed and validated by the authors.
The results showed that when only weather change is considered, the margin of error could be acceptable for some applications
(the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation
1.scenarios,
Introduction
the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered).
The value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the
Water inis the
decrease indispensable for human
number of heating hours life, however
of 22-139h water
during the resources
heating seasonare (depending
declining recently. Nowadays,
on the combination about 768
of weather and
million people
renovation are lack
scenarios of drinking
considered). On thewater
other in the function
hand, world, and thereincreased
intercept will be for
3 billion people
7.8-12.7% live in(depending
per decade water-stressed
on the
countries in the future
coupled scenarios). The2025
values[1]. Even though
suggested could bein used
water-rich areas,
to modify the it’s still difficult
function forfor
parameters people to find clean
the scenarios water and
considered, in
improve the accuracy of heat demand estimations.

© 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and
* Corresponding author. Tel.: +86 551 6290 5590; fax: +86 551 6360 6459.
Cooling.
E-mail address: hwei@hfut.edu.cn
Keywords: Heat demand; Forecast; Climate change
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the 9th International Conference on Applied Energy.

1876-6102 © 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the 9th International Conference on Applied Energy.
10.1016/j.egypro.2017.12.538
1610 Shanshan Liu et al. / Energy Procedia 142 (2017) 1609–1614
2 Author name / Energy Procedia 00 (2017) 000–000

remote outdoors. Great efforts have been made to solve these problems [2, 3], and there were several water
generating methods among which atmospheric vapor processing attracted researchers’ attention. The atmosphere
was estimated to contain over 12.9×10 12 m3 of renewable water, and extracting water from air was potential and
attractive [3]. During the past decades, thermoelectric (TE) technology has drawn attention around the world. TECs
take use of Peltier effect to produce temperature difference when direct current (DC) is applied. Without moving
parts, refrigerant and emission gases, TECs are quiet, reliable and environmental-friendly. The TECs are also small,
light and can be directly powered by photovoltaic (PV) panels. Researchers studied and improved the performance
of TECs from materials, system structure, heat dissipation method and so on [4-6]. Integrating TECs into
atmospheric vapor harvesting is an attractive way to design a portable water generator. Many works investigated
vapor extracting performance by TECs. Suryaningsih et al. [7] studied the influence of large inlet flow rates. Joshi et
al. [8] studied the impact of small flow rates on a water generator with 10 TE modules. Bortolini et al. [1] did
experiment with 20 TE modules at large flow rates. Those works studied large-size water generators with 10-20 TE
modules, which weren’t suitable for outdoor use. And there were few reported literatures studying portable water
generators. Generally, a portable water generator should be small, light and could work at small air flow rates with
less energy consumption, which determined that the number of TECs couldn’t be too much. To provide water for
hikers and scientific investigation teams, this study designed a portable water generator with two TE modules.
Different inlet RH and small air flow rates were experimentally investigated to study their impacts on generated
water and condensation rate. This paper also analyzed some factors influencing the condensation rate and gave some
tips to improve the performance of system.

2. Experiment

2.1. Experimental setup

The experimental system contained a humidifier, a mixing chamber, an air channel and a TE water generator.
The air channel was 1.2 m and insulated by polystyrene foam. The humidifier was linked to the mixing chamber
(0.5625 m3) to control the RH. Air flow rates were controlled by a cross flow fan. The TE water generator was
connected to the end of the air channel. The heat dissipation of TE module used a cross flow fan. The condensed
droplets were collected by a water pan. The schematic diagram of experimental system and TE water generator were
shown in Fig.1. The water generator was 0.46 m×0.14 m×0.15 m, and the weight was 7 kg. The thermal slug of cold
side was 0.18 m×0.1 m, and the height and spacing of fins were 0.02 m and 0.002 m. The total areas of fins were
0.216 m2. The thermal slug of hot side was 0.23 m×0.12 m and the height of fins was 0.03 m.

Fig. 1. (a) The schematic of water generator (b) The photo of experimental rig. 1-Mixing chamber; 2-Inlet channel; 3- Hygrothermogrophs; 4-
Inlet; 5-Thermal slug on the cold side of TECs; 6-Water pan; 7-Thermal slug on the hot side of TECs; 8-Cooling fan; 9-Outlet; 10-Humidifier;
11-DC power; 12-Data acquisition instrument; 13-Electric balance; 14-Anemometer.

2.2. Experimental procedure

Air flowed into the mixing chamber by the cross flow fan and then be humidified. Different RH of inlet air was
controlled by the humidifier. Then the humidified air flowed through TECs by an inlet air channel, and the
temperature of inlet air was reduced by TECs to its dew point. Condensate droplets would grow and finally dropped
 Shanshan Liu et al. / Energy Procedia 142 (2017) 1609–1614 1611
Author name / Energy Procedia 00 (2017) 000–000 3

down to the water pan. Collected water was measured every hour by an electronic balance. Input power source was
controlled as 3.4 A, 15.4 V throughout the experiments. Three different flow rates were controlled by a cross flow
fan, and were measured by an anemometer. Temperature (T) and RH of inlet and outlet air were measured by
hygrothermogrophs. The temperature of fins was measured by thermocouples. A data acquisition instrument was
used to record data.

3. Results and discussion

The results of experiments were shown in Table 1.

Water vapor partial pressure of saturated moist air is [9]:
2 3991.11
{ ps }kPa
 exp[18.5916  ] (1)
15 {T }C  233.84

where Ps was partial pressure of saturated moist air, Pa; T was temperature, °C.
The specific humidity of inlet or outlet air can be calculated by [9]:
 ps
d  0.622 (2)
p   ps

where d was specific humidity of air, kg/kg dry air; φ was RH, %; p was total pressure of moist air, Pa.
The condensation rate η was calculated by:

Generated Water
 (3)
Inlet moisture content
Table 1. Data of condensation experiments.
Exp. Flow rate (m3/h) RH (%) T (°C) Water (g/h) Condensation rate η (%) Water (kg/kWhm2)
1 29.7 67.8 24.3 11.2 2.42 0.889
2 29.7 77.7 23.1 15.2 3.08 1.207
3 29.7 84.7 24.2 22.3 3.87 1.771
4 29.7 92.7 23.6 25.1 4.10 1.993
5 29.7 86.8 21.9 18.2 3.56 1.445
6 50.4 86.5 22.1 18.7 2.13 1.485
7 70.6 86.5 21.2 20.6 1.76 1.636

3.1. The impact of relative humidity

The amount of collected water at different RH was shown in Fig. 2. The amount of hourly collected water was
increased with time went on. After 6 or 7 hours, the amount of collected water gradually became constant and the
average hourly generated water was shown in Fig. 3. This behavior could be explained that the condensate droplets
were small in the beginning, and the disturbance of inlet air and effect of gravity couldn’t offset the adhesion. So the
droplets were adhered on the fins. The condensation rate was shown in Fig. 4. The amount of condensate water and
condensation rate was both proportional to RH. When inlet air temperature, flow rate and input power kept constant,
the increased RH led to rising moisture content. The moisture content was calculated by equation (2), as shown in
Fig. 4. The increased RH also led to the rising dew point temperature of inlet air. As a result, temperature difference
(ΔT) between inlet air and its dew point decreased with rising RH, as shown in Fig. 5. The declining ΔT led to a
1612 Shanshan Liu et al. / Energy Procedia 142 (2017) 1609–1614
4 Author name / Energy Procedia 00 (2017) 000–000

more easy condensation. In summary, increased inlet RH was beneficial and brought increased amount of generated
water and condensation rate.
67.8% RH
25 77.7% RH 26
84.7% RH 24
92.7% RH

Water generation rate (g/h)

20 22
Generated water (g/h)

20
15
18

16
10
14

5 12

10
0
8

0 1 2 3 4 5 6 7 8 9 10 11 65 70 75 80 85 90 95
Time (h) Relative humidity (%)

Fig. 2. Variation of generated water with time at different RH. Fig. 3. Water generation rate at different RH.

0.0175
Generated water/inlet moisture content(%)

0.0170 4.5
Inlet moisture content (kg/kg dry air)

4.0

Temperature difference T (C)

0.0165
4.0
0.0160
3.5
3.5
0.0155

0.0150 3.0

0.0145 2.5
3.0
0.0140
2.0
0.0135
2.5 1.5
0.0130

0.0125 1.0
65 70 75 80 85 90 95 65 70 75 80 85 90 95
Relative humidity (%) Relative humidity (%)

Fig. 4.Inlet moisture content and generation ratio at different RH. Fig. 5. ΔT at different RH.

3.2. The impact of air flow rates

Water production at different flow rates was shown in Fig. 6, and average hourly generated water after
stabilization was shown in Fig. 7. In the beginning, the amount of condensed water at 29.6 m3/h was more than that
of the air flow at 50.4 m3/h and 70.6 m3/h. The high temperature of fins may explained the less water generation
amount at high flow rate in the beginning, as shown in Fig. 8. After 8 hours, the generated water gradually stabilized
and rose with increased flow rates. This was explained by a fact that the temperature of cold side fins decreased with
time went by and the rising flow rate increased the amount of vapor content in the channel.
25 21.0
29.6 m3/h
50.4 m3/h 20.5
20 70.6 m3/h
Water generation rate (g/h)

20.0
Generated water (g/h)

15 19.5

19.0
10
18.5

5
18.0

17.5
0

17.0
0 1 2 3 4 5 6 7 8 9 10 11 20 30 40 50 60 70 80
Time (h) Flow rate (m3/h)

Fig. 6. Generated water at different air flow rates. Fig. 7. Generated water at different air flow rates.
 Author name / Energy Procedia 00 (2017) 000–000 5
Shanshan Liu et al. / Energy Procedia 142 (2017) 1609–1614 1613

18 3.0
29.6m3/h
3.5
50.4m3/h
Temperature of cold side fin (C)
2.5

Temperature difference (C)

16 70.6m3/h

Ratio of inlet moisture (%)

2.0 3.0

14
1.5
2.5

12 1.0

2.0
0.5
10

0.0 1.5
0 100 200 300 400 500 20 40 60 80
Time (min) Flow rate (m3/h)

Fig. 8. Temperature of cold side fin at different air flow rates. Fig. 9. Temperature difference and condensation rate.

From Figs. 7 and 9, the rising inlet air flow rate resulted in increased water amount but declining condensation
rate. On the one hand, increased air flow rate brought more inlet moisture and stronger disturbance. But the system
with two TE modules couldn’t supply enough cooling capacity with increased air flow rate, so the condensation rate
decreased. On the other hand, increased air flow rate also led to insufficient contact time between moist air and
condensate surface. These both led to reduced temperature of cold side fins and temperature difference of inlet and
outlet air, as shown in Figs. 8 and 9, which may accounted for the declining condensation rate.

3.3. Analysis of condensation performance of the system

Performance comparisons of different TE water generators were shown in Table 2. The experimental prototype in
this study (superscript was ) with 2 TE modules performed well with low RH and less energy consumption. And
from the results of papers [1, 10], long air channel or U type air channel could increase the condensation rate
significantly, which was a result of longer contact time and continuous cooling.

Table 2. Performance comparisons of different TEC water generators.

TEC Inlet air parameters System performance
used T (°C) RH (%) Flow rate (m /h) 3
TEC consumed power (W) Water (g/h) Rate (%)
2 
24.29 67.8 29.7 52.3 11.2 2.42
2[7] 25.3 60 129.1 65.5 27.9 1.78
[10]
3 27 82 13.68 100 40.3 13.62
10[8] 30 65 80.3 135.0 5.0 0.31
[1]
20 32 39.5 57 1000 597.4 76.2

Small cooling capacity, small condensation surface and single pass structure were the main reasons for low
condensation efficiency. The wetting property of condensation surface was also an important factor, which was
rarely discussed. The fin spacing of cold side thermal slug in this work was 2 cm, which provided more
condensation surface in the unit volume, but it would produce liquid-bridge between fins. Liquid-bridge formed a
resistance layer and had negative effect on heat transfer and condensation process. Heat transfer rate in dropwise
condensation is more than 10 times larger than film condensation. Typically, on a nonstructured hydrophobic
surface, water drops need to reach 2~3 mm in diameter to slide. Average drop departure size for super–hydrophobic
vertical surface (by modifying the surface topography or chemical composition) could be reduced to 250 μm [11],
which promoted the droplets departure and heat transfer significantly. Modifying the wetting property of
condensation surface was an effective way to promote the amount of generated water.

4. Conclusion and future work

1614 Shanshan Liu et al. / Energy Procedia 142 (2017) 1609–1614
6 Author name / Energy Procedia 00 (2017) 000–000

This study designed a portable water generator with TE method. The system was experimentally investigated at
different inlet air flow rates and RH. The amount of generated water increased with rising inlet RH and inlet air flow
rate and the maximum generated water was 25.1 g/h. The condensation rate increased with rising RH but decreased
with increased air flow rate. Small cooling capacity, small condensation surface, single pass structure and adhesion
of droplets on condensate surface were the main reasons for low condensation efficiency. The water generator could
be supplied by a PV panel directly without operation cost, which was energy-saving. In the future, more efforts need
be done to optimize the system structure to increase contact surface and time between air and cold side of TE
modules. To promote dropwise condensation improving heat transfer and condensation process, the wetting property
of condensate surface also needs modification.

Acknowledgements

This research was supported by the grants from Research on Science and Technology of Anhui Province (No.
1604e0302002), Dongguan Innovative Research Team Program (No.2014607101008), Beijing advanced innovation
for future urban design (UDC2016040200) and also National Natural Science Foundation of China (No. 51606002).

References
[1] M. Bortolini, M. Gamberi, A. Graziani, A. Persona, F. Pilati, A. Regattieri, Air flow optimization for drinking water production through air
dehumidification.
[2] C. Yıldırım, S.K. Soylu, İ. Atmaca, İ. Solmuş, Experimental investigation of a portable desalination unit configured by a thermoelectric cooler,
Energy Conversion and Management, 85 (2014) 140-145.
[3] D. Milani, A. Qadir, A. Vassallo, M. Chiesa, A. Abbas, Experimentally validated model for atmospheric water generation using a solar
assisted desiccant dehumidification system, Energy and Buildings, 77 (2014) 236-246.
[4] D. Zhao, G. Tan, A review of thermoelectric cooling: materials, modeling and applications, Applied Thermal Engineering, 66 (2014) 15-24.
[5] I. Chowdhury, R. Prasher, K. Lofgreen, G. Chrysler, S. Narasimhan, R. Mahajan, D. Koester, R. Alley, R. Venkatasubramanian, On-chip
cooling by superlattice-based thin-film thermoelectrics, Nature Nanotechnology, 4 (2009) 235-238.
[6] M. Russel, D. Ewing, C. Ching, Characterization of a thermoelectric cooler based thermal management system under different operating
conditions, Applied Thermal Engineering, 50 (2013) 652-659.
[7] S. Suryaningsih, O. Nurhilal, I.M. Joni, C. Panatarani, Optimal design of an atmospheric water generator (AWG) based on thermo-electric
cooler (TEC) for drought in rural area, in: AIP Conference Proceedings, AIP Publishing, 2016, pp. 030009.
[8] V. Joshi, V. Joshi, H. Kothari, M. Mahajan, M. Chaudhari, K. Sant, Experimental Investigations on a Portable Fresh Water Generator Using a
Thermoelectric Cooler, Energy Procedia, 109 (2017) 161-166.
[9] J.T. Weidao Shen, EngineeringThermodynamics, Beijing,Higher Education Press, (2007).
[10] J.G. Vián, D. Astrain, M. Domınguez, Numerical modelling and a design of a thermoelectric dehumidifier, Applied Thermal Engineering, 22
(2002) 407-422.
[11] C. Dietz, K. Rykaczewski, A. Fedorov, Y. Joshi, Visualization of droplet departure on a superhydrophobic surface and implications to heat
transfer enhancement during dropwise condensation, Applied physics letters, 97 (2010) 033104.