australian solar cooling interest group (ausscig ... · pietro finocchiaro – csiro, sydney...

Australian Solar Cooling Interest Group (ausSCIG) Conference 2013 www.ausSCIG.org

1

Day 2 – Solar Cooling Conference - 12/04/2013Venue: CSIRO, North Ryde, Sydney

Australian Solar Cooling 2013 Conference

CSIRO, Sydney

11-12 April 2013

Pietro Finocchiaro – CSIRO, Sydney

12.04.2013

NEW DEC OPEN CYCLE FOR AIR-CONDITIONING BASED ON FIXED

COOLED ADSORPTION BEDS AND WET HEAT EXCHANGERS

DEIM -Dipartimento di Energia,ingegneria dell’Informazione, e modelli Matematici

Pietro Finocchiaro, Marco Beccali


2


INTRODUCTION

� Thermally driven open cooling cycles are based on a combination of adsorption processes

and evaporative cooling

� In common DEC air handling units, the core components are desiccant wheel, rotating

sensible heat exchanger and humidifiers


12.04.2013

� System based on simultaneous adsorption and desorption processes

� The use of desiccant rotors implies that condensation heat is rejected

into the processed air and has to be removed by means of the

indirect evaporative process

� The rotating sensible heat exchanger has to carry over two tasks:

� Heat recovery

� Cold production

� Enthalpy difference of the DEC AHU strongly dependent on the

efficiency of sensible heat exchanger and return humidifier

� Leakages and moisture carry over across rotors sealings possible


3


ADSORPTION PROCESS BASED ON DESICCANT ROTORS

� Adsorption process realized by means of desiccant rotors is a quasi – isoenthalpic

transformation

� It presents the disadvantage of causing a temperature increase of the desiccant material

� This phenomena is mostly caused by

the release of adsorption heat due to

water condensation in the desiccant

material and in a lower degree by the

carry-over of heat stored in the

Dehumidification by

desiccant rotor


12.04.2013

tem

pera

tue

absolute humidity

carry-over of heat stored in the

desiccant material from the

regeneration section to the process

section. (The purge section can limit this

last aspect)

� No enthalpy difference, no cold

production

Outside air

Dehumidification by

cooling coil

Enthalpy difference


4



� The higher the

temperature of the

desiccant the lower the

adsorption capacity

� During adsorption in

desiccant rotors, air

50%

60%

70%

80%

90%

100%

RH

[%

] 20 40 60 80

Equilibrium isotherm of

adsorption for silica gel RD

Inlet air

conditions

Air temperature


12.04.2013

desiccant rotors, air

temperature increases

and relative humidity

strongly decreases

0%

10%

20%

30%

40%

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Adsorption bed humidity [kgwater/kgsilica]

Outlet air

conditions


5



Furthermore:

� An increase of the desiccant material temperature is

responsible also for higher regeneration temperatures

required

� Regeneration has to be continuously operated if cold

production is requested

� No opportunity to store adsorption capacity into the

desiccant material since they are built to host a relatively


12.04.2013

desiccant material since they are built to host a relatively

low mass of adsorbent.

� Only option for energy storage is related to the heat

capacity of the regeneration liquid loop

� The use of hot air as regeneration fluid is suitable only

with systems without storage


6


ADSORPTION BED: HEAT AND MASS TRANSFER MECHANISM

� The developed component allows a simultaneous mass transfer between the moist air and

the adsorbent media and heat exchange between the air and the water flowing into the

heat exchanger tubes

� Cooling of the desiccant material during the adsorption process is possible, allowing high

dehumidification performances of the desiccant bed and in general better overall energy

performances of the system

� Water temperatures required can be easily achieved with a cooling tower


12.04.2013


7


ADSORPTION BED: HEAT AND MASS TRANSFER MECHANISM

� Very low humidity ratio can be obtained ;

� Adsorption and desorption processes happen in different times;

� Solar energy can be efficiently stored in the desiccant in terms of adsorption capacity which

can be used later when regeneration heat is not available, strongly reducing the necessity

for thermal storage;


12.04.2013


8


ADSORPTION PROCESS BASED ON ADSORPTION COOLED BED

� The thermodynamic process causes an enthalpy difference between inlet and outlet air

conditions

� Condensation heat can be rejected

tem

pera

tue

� In general, the temperature of air

exiting the adsorption bed is lower

than the one of incoming air

� Downstream indirect evaporative

cooling process can be operated at

Dehumidification by

desiccant rotor

Enthalpy difference


12.04.2013

tem

pera

tue

absolute humidity

cooling process can be operated at

lower temperature Outside air

Dehumidification by

cooling coil


9


� During adsorption in

adsorption cooled bed,

air temperature can

sliglty decrease and

relative humidity

ranges from 15% at the 50%

60%

70%

80%

90%

100%

RH

[%

] 20 40 60 80

Equilibrium isotherm of

adsorption for silica gel RD

Inlet air

conditions

Outlet air

ads cooled

bed

ADSORPTION PROCESS BASED ON ADSORPTION COOLED BED

Outlet air

Air temperature


12.04.2013

ranges from 15% at the

beginning of the

process to 40% at the

end

� Average adsorption

bed humidity at

equilibrium is much

higher than for

desiccant rotors

0%

10%

20%

30%

40%

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Adsorption bed humidity [kgwater/kgsilica]

Outlet air

desiccant

rotor


10


DESIGN OF THE COMPONENT

� The total amount of silica gel has to be chosen

according to the dehumidification rate requested,

flow rate to be processed and desired storage of

adsorption capacity

� In order to minimize pressure drops across the bed

the most important issues are:

The adsorption bed should be designed taking into account the following issues:


12.04.2013

the most important issues are:

� High rate of empty spaces between the

desiccant grains (global porosity of the bed)

� Low air velocity

� Large crossing area, minimum length

� Pressure drop can be similar or lower to the ones

typical for rotors


11


MEASUREMENT TESTS AT UNIPA ON ADSORPTION BEDS

Adsorption bed with cooling

AdsorptionDesorption

60

70

80

20

25

x in x out T in

T out T w out T w in

60

70

80

20

25

x in x out T in T out

Cross section area: 0.24 m 2

Weight of silica gel: 18 kg

Lenght: 13 cm

Flow rate: 210 m3/h

V: 0.26 m/s

DP: 105 Pa


12.04.2013

0

10

20

30

40

50

0

5

10

15

0:00 1:00 2:00 3:00 4:00 5:00 6:00 7:00 8:00

[°C

]

[g

/kg

]

[h]

0

10

20

30

40

50

0

5

10

15

0:00 1:00 2:00 3:00 4:00 5:00 6:00 7:00 8:00

[°C

]

x [

g/

kg

]

[h]


12


INDIRECT EVAPORATIVE COOLING BASED ON WET PLATE HEAT EXCHANGERS

� cross flow plate heat exchanger operated under

wet conditions

� the surface of secondary flow air channels is

wetted by water spraying, such that the water

evaporates into the cooling air and decreases the

temperature of the heat exchange surface


12.04.2013


13


45

50

40

45

50

� Saturation inside the heat exchanger possible

� Temperature of secondary air is close to the

local wet-bulb temperature of the air stream

which increases gradually during the

humidifying process

INDIRECT EVAPORATIVE COOLING: COMPARISON BETWEEN THE SOLUTIONS

� Saturation inside the heat exchanger

not possible

� Secondary air flow passing through

the channels rapidly increases its

temperature


12.04.2013

10

15

20

25

30

35

40

6 7 8 9 1011121314151617181920

tem

pera

ture

humidity

10

15

20

25

30

35

40

6 7 8 9 1011121314151617181920

tem

pera

ture

humidity

Efficiency: 67% Efficiency: 76%


14


10

12

14

16

65

70

75

80

85

Efficiency T2 in = 25°C Efficiency T2 in = 30°C

Cooling Power T2 in = 25°C Cooling Power T2 in = 30°C

MEASUREMENT TESTS ON WET HEAT EXCHANGERS

87%97%

68%

89%

100%

70%

80%

90%

100%

[Co

oli

ng

po

we

r]

T in HX 1 = 40°C wet operation

T in HX 1 = 40°C dry operation

Efficiency and cooling power Performances vs mass flow rate ratio


12.04.2013

0

2

4

6

8

10

30

35

40

45

50

55

60

65

30 40 50 60

[kW

]

[%

]

T1 in [°C]

56%

33%

68%

0%

10%

20%

30%

40%

50%

60%

0% 20% 40% 60% 80% 100%

[Co

oli

ng

po

we

r]

Mass flowrate ratio [%]

Flow rate: 1200 m3/h

Flow rate ratio : 1/1

Flow rate of primary:

1200 m3/h


15


DESIGN CONCEPT OF THE NEW DEC CYCLE

Adsorption bed

Dehumidification +

1 st step cooling

Wet heat exchanger

2 nd step coolingOutside

air

Return air

Exausted air


12.04.2013

� System based on a combination of the proposed solutions for air dehumidification and cooling

� Avoiding or minimizing the use of auxiliary energy source

� System designed for air ventilation, dehumidification and cooling (heating in winter is also

possible)

� Dehumidification and regeneration operated using outside air

� Regeneration carried out using solar air collectors

Exausted air


16


� A portion of the primary air flow rate exiting the wet heat exchanger is drown into the

secondary side

� Electricity consumptions of the system are related to the use of two fans, two pumps and a

cooling tower

DESCRIPTION OF THE NEW DEC CYCLE

� System based on the use of two fixed packed desiccant beds of silica gel operating in a batch

process and cooled by cooling tower, and two wet evaporative heat exchangers connected in

series

� A system of air dumpers provides the commutation between the two adsorption beds in order

to guarantee a continuous dehumidification process

� No auxiliary device included

7

55

60

6540% of supply

air flow rate

50% of supply

air flow rate

100% of air

flow rate


12.04.2013

12

3

4

7

5

8

6

10

15

20

25

30

35

40

45

50

55

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

Te

mp

era

ture

[°C

]

Humidity ratio [g/kg]

40% of supply

air flow rate

100% of air

flow rate


17


TRNSYS SIMULATIONS

� For the adsorption bed a semi-empirical model (Type 165) partially based on the

approach suggested by Pesaran and Mils was created

� For the simulation of the wet heat exchanger, a modification of the Type 757 was

necessary in order to correctly assess the influence of the variation of the mass flow

rate ratio between the primary and secondary side of the heat exchanger.

� Simulation time step is one minute in order to efficiently describe the dynamic effect

of the desiccant beds

� Commutation time between adsorption and desorption phases is fixed and equal to

0.25 h.


12.04.2013

0.25 h.

� The system has the configuration described above and is connected to a conditioned

space of about 1200 m3 used as a classroom

� Occupation profile is 150 persons from 8:00 to 13:00, 20 from 13:00 to 15:00 and

100 from 15:00 to 18:00

� The weather data file used is related to the location of Palermo (South Italy) 38°6' N

13°20' E.


18


TRNSYS SIMULATIONS

� AHU flow rate delivered to the conditioned space is 8300 m3/h

� Maximum sensible and latent loads including internal gains, ventilation and infiltration,

are respectively 25 and 51 kW

� Each adsorption bed has a total volume of 0.3 m3 and contains about 115 kg of silica gel

in grains

� A surface of 46 m2 of solar air collectors provides the heat necessary for the

regeneration of the desiccant material

� Solar collector flow rate is 4200 m3/h.

� The maximum electricity consumption for the fans and the pumps is approximately 4.5


12.04.2013

� The maximum electricity consumption for the fans and the pumps is approximately 4.5

kW when operated at full speed

� The cooling power of the AHU can be varied controlling the speed of the process air fan

� Five days of operation (19th - 23th July) to focus on instantaneous system performances

during typical hot and humid summer days


19


150

200

250

300P

ressu

re d

rop

[P

a]

0.6 - 0.46 m/s

0.5 - 0.46 m/s

0.4 - 0.46 m/s

0.6 - 0.35 m/s

MODEL OF THE ADSORPTION BED: PRESSURE DROPS VERSUS LENGTH, AIR

VELOCITY AND POROSITY


12.04.2013

0

50

100

150

0.1 0.2 0.3

Pressu

re d

rop

[P

a]

Lenght [m]

0.5 - 0.35 m/s

0.4 - 0.35 m/s

0.6 - 0.28 m/s

0.5 - 0.28 m/s

0.4 - 0.28 m/s


20


ADSORPTION BED VS DESICCANT WHEEL: COMPARISON OF PRESSURE DROPS AT

EQUAL AVERAGE DEHUMIDIFICATION RATE

T in pro 23 °C Area pro 0.1364m2

x in pro 16 °C Flow rate 1250 m3/h

T out pro 55 g/kg Flow mass ratio 2/3

x out pro 5.9 g/kg T rig 80 °C

dx pro 10.1 g/kg Air velocity 2.55 m/s

Dp 188.0 Pa

Desiccant rotor typical operation data

Specific dehumidification capacity: 0.081 kg/h/Pa 0.80

0.6 0.5 0.4 Desiccant rotor


12.04.2013

Adsorption bed

Specific dehumidification capacity: 0.081 kg/h/Pa

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.1 0.2 0.3

[kg

/h/P

a]

Length [m]

Inlet air condition are the same

Adsorption bed cross area: 1 m2

Air velocity: 0.41 m/s

Mass of silica gel: 78 kg for porosity 0.6 and length 0.1 m

350 kg for porosity 0.4 and length 0.3 m


21


TEMPERATURE AND HUMIDITY OF THE AIR IN THE CONDITIONED ROOM

Five days of operation (19th - 23th July)

Building

temperature

Building relative

humidity

Outside humidity

ratio

Outside

temperature


12.04.2013

Flow rate


22


TEMPERATURES IN THE AHU COMPONENTS


Outlet

temperaure of

solar collectors

Supply

temperature

Range

17-23°C


12.04.2013

Supply wet bulb

temperature


23


COOLING POWER OF THE AHU, SOLAR HEAT PRODUCTION, EFFICIENCY OF

WET HEAT EXCHANGERS


Cooling power

Wet heat

exchanger

efficiency

Solar heat

Power

Range 60

- 65 %

Range 50 -

90 kW


12.04.2013


24


Daily average data

T building [°C] 26.4

RH building [%] 55.2

T supply [°C] 20.7

x supply [g/kg] 10.3

Monthly data - July

SIMULATION RESULTS OF TRNSYS MODEL

3146,

2550,

24%

P Ads Beds P HX 1 P HX 2


12.04.2013

Operation hours [h] 341

Cooling energy delivered by the DEC system [kWh] 10360

Electricity consumed by the DEC system [kWh] 1210

Heat delivered by the solar collectors [kWh] 7450

Electrical COP [-] 8.6

Thermal COP [-] 1,7

Water consumption [m3] 12.1

4985,

47%

3146,

29%


25


CONCLUSIONS

� An innovative solar DEC concept using a combination of fixed cooled adsorption beds

packed with silica gel and two wet heat exchangers was developed and analyzed

� The proposed adsorption bed realizes the simultaneous mass and heat transfer, permitting

to dehumidify and to cool the processed air

� Simulation results have shown high dehumidification potential even in presence of high

humidity values and the opportunity to store the adsorption capacity

� A lower air velocity is needed in comparison to the standard values commonly used for AHU

components

� An accurate design of the desiccant bed can limit pressure drops under 100 Pa


12.04.2013

� An accurate design of the desiccant bed can limit pressure drops under 100 Pa

� Performance tests have been carried out also on wet plate heat exchangers in several

controlled air conditions, with the aim to assess the efficiency for unbalanced flow rate ratio

� A complete TRNSYS model of the proposed DEC system was created

� Simulation results show that high electrical and thermal COPs can be expected in

comparison to traditional DEC systems


26


PROTOTYPE OF A COMPACT SYSTEM FOR RESIDENTIAL APPLICATION

Collector area: 2 m 2

Flow rate: 500 m 3/h

Max cooling power: 3 kW

Max cooling power to the

Patent pending


12.04.2013

Max cooling power to the cond. room: 1.8kW


27


Thank you for your attention!


12.04.2013


28


PERFORMANCE OF THE ADSORPTION BED IN HUMID CLIMATES

20

25

30

35

40

0.015

0.020

0.025

0.030

C]

[kg

/kg

]

Cross section area: 1m 2

Empty space rate: 0.5

Lenght: 0.2 m

Tin: 30°C

x: 22 g/kg

Initial water content of bed: 0.05 kg/kg

M water in : 2000 kg/h


12.04.2013

0

5

10

15

20

0.000

0.005

0.010

0.015

0.2

5

0.5

0.7

5 1

1.2

5

1.5

1.7

5 2

2.2

5

2.5

2.7

5 3

3.2

5

3.5

3.7

5 4

[°C

]

[kg

/kg

]

[h]

x in x out

T in T out

water in

T water in : 25 °C

DP: 85 Pa


29


PERFORMANCE OF THE ADSORPTION BED IN HUMID CLIMATES

Cross section area: 1m 2

Bed porosity: 0.5

Lenght: 0.2 m

Mass of silica: 194 kg

Tin: 30°C

x: 22 g/kg

Initial water content of bed: 0.05 kg/kg6.0

8.0

10.0

12.0

14.0

70

75

80

85

90

[kW]

[kJ/

kg

K]


12.04.2013

M water in : 2000 kg/h

T water in : 25 °C

DP: 85 Pa

0.0

2.0

4.0

6.0

50

55

60

65

0

0.2

5

0.5

0.7

5 1

1.2

5

1.5

1.7

5 2

2.2

5

2.5

2.7

5 3

3.2

5

3.5

3.7

5

[kJ/

kg

K]

[h]

h in

h out

Cooling power


30


0.150

0.200

0.250

0.300

[g

/kg

/P

a]

Desiccant rotor

0.5 - 0.3 m

0.5 - 0.2 m

0.5 - 0.1 m

ABSOLUTE HUMIDITY DIFFERENCE FOR UNITARY PRESSURE DROP


12.04.2013

0.000

0.050

0.100

0.150

0.2

5

0.7

5

1.2

5

1.7

5

2.2

5

2.7

5

3.2

5

3.7

5

[g

/kg

/P

a]

[h]


31


COMPARISON BETWEEN TRNSYS MODEL AND MEASUREAMENT DATA



Lenght: 0.13 m



12.04.2013


32




Lenght: 0.13 m




12.04.2013


33




Lenght: 0.13 m




12.04.2013

australian solar cooling interest group (ausscig ... · pietro finocchiaro – csiro, sydney...

Documents