Alternative Non - CFC Mobile Air Conditioner

8/11/2019 Alternative Non - CFC Mobile Air Conditioner

1/100

3 4456,

362925

0

OAK RIDGE

NATIONAL

LABORATORY

I

~

~

TIN MARIETTA ENERGY S

THE UNITED STATES

ORNUCON-335

ALTERNATE NON-CFC

MOBILE

AIR

CONDITIONING

V. C. Mei, F. C. Chen, and D.

M.

Kyle

Energy Division

O A K R I D G E N A T I O N A L L A B O R A T O R Y

C E N T R A L R E S E A R C H L I B R A R Y

C I R C U L A T I O N

SECT ON

4500N ROOM 1 7 5

L I BRARV

LOAN

COPV

D O N O T T R A N S F E R T O A N O T H E R P E R S O N

I f y o u

w i s h s o m e o n e e l s e to see t h i s

r e p o r t , s e n d i n n a m e w i t h r ep o r t a n d

the l i b r a r y w i l l a r r a n g e o l o a n .

September 1992


2/100

Available to DOE and

DOE

contrectors from

the OIfiCe

of

Scienmzc and

T e

cal Information, P.O. Box 82, Oak Ridge-TN

37831; pricee available

from

(615)

5768401, FTS 626-6401.

Availabbe to

the

public

from

the Nationel

Techniil

lnfafllleoion Service U.S.

Department of Comunerce.5285 Port Royal Rd.,SprhgfW, VA 22161.


3/100

OMB

Control No

F1332.15

U.S.

DEPARTMENT

OF

ENERGY

0-90)

DOE Repor t No.

191 0-1400

Burden Oisclosure

Statement

on

back

DOE AND MAJOR CONTRACTOR RECOMMENDATIONS FOR

other editions

ANNOUNCEMENT AND DISTRIBUTION

OF

DOCUMENTS

See Instructions on

Reverse

Side

I

3.

DOE

Funding Off ice

2. DOE

Contract

No.

I

4.OSTl UC

or

C Category

No.

Subm it ted by (Name and Posit ion) (Please pr int or type)

DE-AC05 -840R2 400

I

EC 01 04 00 0 I UC-330

Title

.

Phone

574-6764

Typ e of Document ( x one)

a. Scientific and technical repo rt:

D

monthly D quarterly

Cl

annual

D

final

XX

topical speciaVpublic interest

D

other

b. Conference paper: Nam e of conference (no abbreviations)

Organizat ion

Location (city/st/ctry)

Date (mo/day/yr) Sponsor

Contents:

D

proceedings viewgraphs paper @ poster sessions

Document is accompanied by

c. Computer Media:

D magnetic tape(s)

D diskette(s)

D

other (specify)

d. Other (e.g., journal article, manu script, manu al, etc.) (specify)

Signature Date

Copies Transmit ted to

OSTl ( x

one or m ore). See instruct ions on Reverse.

Cl

a. Copies for unclass ified distribution.

Cl c. Twenty-five additional copies for special/public interest.

D

d. Two reproducible copies for OSTl processing.

Cl Yes (x one or more)

D

a. Classified

D

b. Export Control/lTAR/EAR

D

c. Proprietary Data

Cl d. Small Business Innovation Research (SBIR)

e. Unclas sified Controlled Nuclear Information (UCN I)

Cl

f. Patent Hold

D

g. Translations of copyrighted material

D h. Cooperative Research and Development Agreement (CRADA)

Cl i. Other (explain)

Upon demand , OS Tl will make distribution of these documents in accordance with existing laws, regulations, and/or written prog ram

off ce guidance.

Patent Inform at ion

Cl

e. Copies for spe cial distribution.

17 . Copies for classified distribution as defined in M -3679 .

D g. One copy for OSTl classified processing.

b .

Twelve copies for OSTl processing and N TlS sales.

Docum ent conta ins in format ion perta in ing to any of the fo l lowin g:

&N O (If no, unlimited distribution may be made within the U S . In addition, copies may be made available to

foreign requestors through exchange agreeme nts or NTlS sales.)

(Announc e to appropriate recipients as listed in M -3679, Standard D istribution for Class ified Scientific and Technical Reports)

(Boxes b-g will be announced to Government agencies and their contractors)

(This designation mu st be in accordance with written DOE Program O ffice guidance. Attach copy of guidance.)

a. D oes this information product disclose any ne w equipment, process, or material?

b. Has an invention disclosure been subm itted to DO E covering any aspect of this information product? XXN o @ Yes

c. Are there any patent-related objections to the release of this information product?Xi No

11

Yes If so, state these objections.

No

@

Yes

.

If so, identify page nos.

If

so,

identify the DO E (or other) disclosure nu mber and to whom the disclosure was sub mitted.

. Addit ional in for mat ion, remarks, and specia l handl ing ins truct ions . ( D o

not

ident i fy Sigma categories for Nuclear Weapon

Data reports, and d o

not

provid e addit ional instruc t ions which are inconsistent with i tem 8 above.) (Contin ue on separate

sheet, if necessary.)


4/100

AlternativeNon-CFC

obile

Air

Conditioning

BY

V. C. Mei, F. C. Chen and D.

M.

Kyle

Energy Division

Oak Ridge National Laboratory

ORNL/CON-335

6

9b

Prepared

for

Office of Transportation Technologies

U.S.

Department of Energy

September 1992

Prepared by the

OAK RIDGE NATIONAL LABORATORY

Oak Ridge, Tennessee 37831

managed by

MARTIN M AR IElT A ENE RGY SYSTEMS, INC.

for the

U.S.

DEPARTMENT OF ENERGY

under

Contract No. DE-AC05-840R21400

3

4 4 5 6

0362725 0


5/100


6/100

TABLEOF coNTEN s

L I S T O F F I G U R E S

........................................................ v

L I S T O F T A B L E S ......................................................... vii

ACKNOWLEDGMENTS

....................................................

ix

ABSTRACT .............................................................. xi

1. I N T R O D U C I I O N .................................................... 1

2. L I T E R A T U R E S E A R C H ...............................................

5

3. M A C C O O L I N G L O A D

................................................

9

4. A L T E R N A T I V E R E F R I G E R A N T S ....................................... 13

4.1 R-134a

..........................................................

13

4.2 TERNA RY BLENDS

..............................................

13

4.3

OTHER REFRIGERAN T BLENDS

..................................

14

4.4 NON-INERT (FLAMMABLE) RE FRIGERAN TS

........................

14

5. NON-CFC ALTERN ATIVE MAC SYSTEMS ............................... 15

5.1 WOR K ACTUA TED MAC SYSTEMS ................................. 15

5.1.1 Refrigerant Vapor Compression Cycle: Hermetic Systems

.............. 15

5.1.2 Reversed Brayton Air Cycle

....................................

17

5.1.3 Rotary Vane Compressor Air Cycle

..............................

20

5.1.4

Therm oelectric Cooling System

.................................. 21

5.1.5 Stirling Cycle C ooling System .................................... 24

5.3

HEA T ACTUATED MAC SYSTEMS

.................................

26

5.3.1 Ejec tor C ooling System

.......................................

26

5.3.2 Absorp tion C ooling System

.....................................

28

5.3.3

Adsorp tion Cycle (Desiccant) Cooling System .......................

31

5.3.4 Metal Hydride Cooling System .................................. 35

5.2 COMPARISON O F W ORK-ACTUATED MAC SYSTEMS

................

25

5.4 COMPARISON O F HEAT-ACTUATED MAC SYSTEMS ................. 39

6

.

RECOMMENDATIONS FOR FU TURE MAC RESEARCH

AND

DEVELOPMENT ..................................................... 41

6.1 N E A R - T E R M R & D

...............................................

41

6.2 M I D - T E R M R & D

................................................

41

6.3

L O N G - T E R M R & D ...............................................

41

7. CONCLUSIONS ...................................................... 43

8

. R E F E R E N C E S

.......................................................

45

...

111


7/100

TABLEOF

CONTENTS

(continued)

APPENDIX A. OMPUTER CODE FOR BRAYTON CLOSED AIR CYCLE

AND

SAMPLEINPUTDATA

...............................................

49

APPENDIX B

. RNL

MATHEMATICAL MODEL FOR TEMAC

SYSTEM . . . . . . . . .

55

APPENDIX C. RNL COMPUTER CO DE FOR EJECTOR COOLINGSYSTEM . . . . .

69

APPENDIX D . AMPLE CALCULATION OFA METAL HYDR IDE COOLING

SYSTEM

............................................................

79

iv


8/100

.

LISTOFFIGURES

1 Electric compressor drive system ........................................ 16

2a

.

Schem atic of Brayton open-cycle system

................................... 17

2b. Pressure-volume diagram of Brayton air cycle ............................... 18

2c

. Schematic

of

regenerative Brayton closed-cycle system ........................

18

Tem perature-entropy diagram of regenerative Brayton closed-cycle system

.........

d.

19

3. ROVAC mobile air conditioning system ................................... 20

4a

.

Schematic

of

thermoe lectric mobile air conditioning system

.................... 22

4b. Schematic of thermoe lectric liquid-air he at exchanger

.........................

22

4c. Coefficient

of

performance of thermoe lectric mobile air conditioning system ....... 23

5

.

Schematic

of

Stirling mobile air conditioning system ..........................

24

6a. Schematic of ejector cooling system

......................................

26

6b

Schematic of ejector mobile air conditioning system in engine cooling and air

conditioningmode

...................................................

27

7. Schematic of single-stage abs orption cooling system .......................... 29

8

. Absorp tion system cooling capacity

vs

car s p e e d

.............................

30

9. Schem atic of truck absorption refrigeration system ........................... 30

10 . chematic

of

truck absorption refrigeration system components

.................

31

11

.

12a

.

Schem atic of closed-cycle desiccant m obile air conditioning system ...............

Schematic of desiccant mobile air conditioning component arrangement ...........

33

34

12b. Schematic of desiccant bed design ....................................... 34

13 Adsorption isotherms of typical desiccant materials ...........................

35

1 4 a

14b.

Schematic

of

metal hydride cooling system, regeneration mod e

..................

Schematic of metal hydride cooling system. absorption mode

36

36

..................

V


9/100

LIST

OF FIGURES (continued)

14c

Pressure-temperature diagram

of

metal hydride cooling system .................. 36

15.

Schematic

of

metal hydride mobile air conditioning system for hydrogen-fueled

vehicle

............................................................

37

16. Hydriding alloy

...................................................... 38

vi


10/100

LIST

OFTABLES

1

2

3

4

5.

6

.

7

8

.

9.

GWPs

for fluorocarbons and othe r trace greenhouse gases

.....................

2

Predicted automotive cooling load requirements .............................

9

Typical glazing characteristics

........................................... 9

Windowarrangements

................................................ 10

Comparison of calculated ATs........................................... 11

Performance comparison

of

alternative refrigerants

.......................... 14

Reversed Brayton cycle mobile air conditioning calculated performance ........... 19

Calculated and actual measured performance data of

ROVAC

model

30B/45/9.5-4

...

21

Thermoelectric mobile air conditioning coefficient

of

performance

as

a function of

cooler inlet heat transfer fluid temperature .................................

23

10.

Calculated performance

of

Stirling Thermal Motor

STM4-35 SAC

model Stirling

mobile air conditioning system

..........................................

25

11

Comparison

of

work-actuated mobile air conditioning

systems .................. 25

12

Comparison of conventional mobile air conditioning and ejector mobile air

conditioning ........................................................ 28

13

Summary

of

advanced absorption cycle analyses for

80.000

Btu/h output

...........

32

14

Comparison of hea t ac tuated-mobile air conditioning systems ................... 39

I


11/100

i.


12/100

The

authors are indebted

to

Rogelio Sullivan, Office of Transportation Technologies,

Department of Energy, for his support of this

effort.

The authors are also grateful

to

Richard W.

Murphy for providing the computer codes for the Brayton cycle cooling systems and ejector cooling

systems work. Finally, we would like to thank Do nna Penland for her

skill

and patience in preparing

the manuscript.

ix


13/100


14/100

Concern about the destruction of the global environment by chlorofluorocarbon

(CFC)

fluids

has become an im petus in the se arch for alternative, non-CFC refrigerants and cooling methods for

mobile air conditioning (MAC). While some alternative refrigerants have bee n identified, they a re

not considered a lasting solution because

of

their high global warming potential, which could re sult

in their eventu al phaseout. In view of this dilemm a, environmentally acceptab le altern ative cooling

methods have become important. This report, therefore,

is

aimed mainly at the study

of

alternative

automotive cooling methodologies, although it briefly discusses the current status

of

alternative

refrigerants.

Th e alternative

MACs

can be divided into work-actuated and heat-actuated systems. Work-

actu ated systems include'conven tional MAC, reversed Brayton air cycle, rotary vane com pressor air

cycle, Stirling cycle, thermoe lectric ( E)cooling, etc. Hea t-actuated MACs include metal hydride

cooling, adsorption cooling, ejec tor cooling, absorption cycle, etc. While we ar e bette r experienced

with som e work-actuated cycle systems, heat-actuated cycle systems have a high potential for energ y

savings with possible waste heat applications. In this study, each alternative cooling method is

discussed for its advantages and its limits.

x


15/100

I


16/100

1.

INTRODUCI'ION

While mobile air conditioning (MA C) is still an op tional item, its marke t sha re has reached

90%

in passenger cars and 70% in light trucks in North America. It is estimated that MAC accounts for

over 30% of all R-12 consumption in Japan (Ushimaru 1989) and 33% in th e United State s (Statt

1988).

MAC accounts for about 20 gal.

of

fuel per 10,000 miles driven in th e U.S. (Ha lbersta dt

1991). T he re ar e about 184.4 million passenger vehicles in the U.S. (Davis and Hu 1991) with an

annual driving average

of

10,119 miles pe r car. With 90% of these equipped with MAC systems,

annual fuel consumption will b e c lose to 1.68billion gal of refined fuel if MAC is used for half of the

annual mileage driven. Efficient MAC systems, judging from the above figures, ar e needed. Th ere

is also a significant potential for growth in the MAC field as the use of private

cars

and trucks

becomes m ore common in Asia, Africa, and So uth and Central America.

Because of the scheduled phaseout

of

R-12 by the year 2000

(Montreal Protocol

1987), MAC

alterna tive refrigerants and cooling me thods are urgently needed. While R-134a has bee n identified

as

the likely replacement for R-12, it is basically for new car models only. Fo r existing cars, some

nonaze otropic refrigerant mixtures have been identified as suitable alternatives. T h e existing M AC

alternative refrigerants may only be short-term solutions, however, because

of

their high global

warming potential (GWP). Table 1 shows the lifetime equivalent CO, emissions for the MA C 500-

year GWP (Fischer e t a].). T he table indicates that th e GWP

of

R-134a is about 420 times higher

than that of CO,. Eithe r a different class of refrigerank o r alternative cooling technologies, both of

which a re environmentally safe, will be needed.

A literature search for MAC-related research and development (R&D) work for the past 20

years was performed to look into the history of MAC evaluation. Th e search provides some

intriguing information that indicates that novel alternative MA C cooling technologies we re never

seriously studied, despite the high potential for saving energy for som e of the concepts, and that most

of th e work was performed before 1985. Since 1987, almost all M AC R & D work was related t o R-12

replacement refrigerants because

of

the Montreal Protocol.

A

vigorous and massive effort that

investigated M AC alternative refrigerants, material compatibility, lubricant oils, and redesigns of the

major MAC components, etc., indicates that MAC industry is racing with time

to

come u p with an

environmentally acceptable MAC system with performance com parable to or b etter than conventional

MAC systems before the year 2OOO. Alternative cooling technologies have more or less been

abandon ed for the time being, even though they have th e potential to provide environmentally sound

energy saving MAC ystems. In this

study,

the recent development

of

MAC with alternative

refrigerants and of some promising alternative cooling technologies, both work-actuated and heat-

actuated, are discussed in detail.

Fo r alternative refrigerants, R-134a has been identified as the most promising replacem ent for

R-1 2 in this report. T he technical problems facing R-134a in retrofitting existing cars ar e discussed.

Some nonazeotropic refrigerant mixtures, such

as

the ternary blends, are discussed for their

advan tages an d limits in re trofitting t he existing cars.

Alternative MAC cooling methods ar e generally divided into work-actuated and heat-actuated

systems. Wo rk-actuated systems ar e those actuated by engin e shaft power or electricity gen erated

throug h th e vehicle electric generating system. Promising work-actuated MAC systems include

hermetic cooling systems, reversed Brayton air cycle, rotary-vane compressor air cycle, Stirling cycle,

1


17/100


18/100

TE cooling, etc. Even though we a re well versed in som e work-actuated systems, such as herme tic

systems, which have been routinely used for residential applications, some technical ch allenges remain

when these systems ar e used for automotive cooling applications. Each on e of them is discussed for

its merits and for technical obstacles to be overcome. Because

of

the possibility

of

waste heat

application for heat actuated MAC cooling systems, their energy saving potential is very good.

However, research and development work in this type of MAC is very rare, which presents a great

challenge in technology assessment and potential developm ent. In this study, ejecto r cooling,

ads orption (desiccant) cooling, absorption cooling, metal hydride cooling, etc., a re identified as the

potential candidates for MA C applications. Each o ne has un ique features and technical limits that

ar e discussed in this study. Som e information on previous work that can be applied to

MAC,

uch

as the Oak Ridge National Laboratory (ORNL) computer codes for ejector cooling, the reversed

Brayton cycle, and th e mathematical model for th e thermoelectric (TE) MA C system, is included in

the appendices.

3


19/100

. .

..


20/100

2 LITERATURESEARCH

A literature search was performed o n M AC systems from 1970 to 1990, and 184 articles were

identified. Persona l contacts, however, proved to be

a

rich resource for updated information. This

search was aimed at finding past work on alternative refrigerants for R-12 and alternative cooling

methodologies for MAC systems.

Spauschus (1988) studied th e com pressor and refrigeration system requirements a nd information

gaps for R-134a application as an R -12 substitute. His main concerns we re th e chemical stability of

R-l34a, th e lack of a reliable detection method for low concentrations of R-l34a, and its solubility

with lubricants. A pap er by Bivens et al. (1989) discusses in detail the us e of ternary blends, such as

R-22Dt-152a/R-124 (o r R-114), as MA C refrigerants. Th ese au thors claim that all refrigerants used

in th e blends a re comm ercially available and that the thermodynamic properties

of

the blends are

close to those of R-12. However, they also mention that th e blends are not a drop-in

type

of R-12

replacem ent. In fact, som e modifications ar e need ed in retrofitting the existing cars, and they could

be quite costly (Dieckmann and Bentley). While th e report by Dieckmann and B entley mainly

discusses the cost of retrofitting existing cars with R-134a refrigerant, it suggests some component

changes that are applicable to the ternary blends

as

well.

An

article by Bate ma n

(1989) discusses

t he

transport and thermodynamic p roperties

of R-134a that re quire special attention.

He

concludes that

a significant amount

of

developm ent work and system redesign will be required to optimize the use

of

R-134a for MAC. A more recent paper by Bateman

(1990)

presents the status of R-134a for

MAC. In this study, the environmental impact, refrigerant properties, heat transfer, refrigeration

performance, material compatibility, and lubricants for R-134a ar e discussed in detail.

El-Bourini, Hayashi, and Adachi (1990) investigated M AC performance with R-134a. Th ese

authors finding was that the se rpentine condenser ne eded to be changed t o a parallel-flow condenser.

Provost and A rrieta (1990) studied the differences between R-12 and R-134a in M AC, which led to

new methods of presenting refrigerant data by using com puter graphics technology. A paper w ritten

by Bateman et al.

(1990)

discusses th e field test results of the ternary blends used for MAC. The se

autho rs summarize the requirements for retrofit

of

the ternary blends

to

existing vehicles. Struss,

Henk es, and Gabbey (1990) experimentally studied both serpentine and parallel condens ers for R-12

and R-134a. They found that R-134a generated a higher head pressure. Improvement

of

condenser

perfo rma nce and refrigerant expansion devices could redu ce head pre ssure for R-134a. El-Bourini,

Adachi, and Tajima (1991) experimentally investigated the effect

of

an expansion valve on R-134a

MAC

system. They found, from wind tunnel and road tests, that better cooling performance and

similar discharge pressures at a lesser refrigerant charge of R-134a could be achieved with a new

expansion valve and parallel-flow condenser than with conventional

R-12

M A C

systems.

Guntly (1990) investigated the use of R-22 as a MAC refrigerant. He concluded that R-22

would n ot degrade MAC system performance and would req uire th e sam e level of redesign required

by R-134a. It

is

clear that most M AC alternative refrigerant studies were don e afte r 1988, which

shows that alternative refrigerants for

MAC

were vigorously investigated only af ter th e Mo ntreal

Protocol called for the scheduled phaseout of R-12 and some other CFC fluids by th e year 2,000.

Bessler an d Fo rbes (1987) studied an electrically driven hermetic M AC system with a brushless

dc motor. While their system could reduc e refrigerant leakage from th e compressor shaft seal, its

main purpose was to save energy. Because of the smaller car engines currently used for better

mileage, an electrically driven M AC system with variable-speed capability could achieve con tinuous

5


21/100

control

of

cooling output independent

of

th e engine drive belt. This design could outp erform an

M A C system with mechanically controlled compressor displacement. Ak aban e e t al.

(1989)

evaluated

an electrically driven herm etic M AC system with

a

scroll compressor. They found tha t the system

coefficient of performance (CO P) was not as high

as

that of conventional

MAC

systems. Extensive

change in vehicle electrical system is also needed. A paper by Ikeda, Yashii, and Tamura (1990)

discusses a herm etic MA C system for electric vehicles. All findings indicated tha t fo r a h ermetic

electrically driven M A C to work, vehicle cooling load would have t o

be

reduced through window

glazing and reduction of fresh air intake. A report prepared by Garrett, Inc., (1977) discussed the

possible application

of

th e Brayton air cycle for M AC ope ration. With realistic assumptions, th e COP

of a Brayton op en air cycle system is around 1.0, which is low compared with that

of

conventional

MAC ystems. Brayton air cycle systems have bee n routinely used for aircraft air-conditioning

purposes du e to their light weight and compactness

(ASHRAE W A C Application Handbook

1991).

T he ro tary vane com pressors air cycle (Edwards 1975) employs a rotary van type com pressor for air

compression and expansion that could be used for MAC. How ever, little attenti on has been paid t o

this cycle by most researchers because of its high pow er consumption.

Th ere a re many publications abou t th e application of TE cooling, bu t

few

of them ar e for MAC.

Stockholm, Pujol-Soulet, and S terna t (1982) experimentally investigated TE cooling for a passenger

railway coach beca use

of

this cooling systems low ma inten anc e requirem ent. Bec ause

of

th e lack

of

a major breakthrough in TE materials for the past 30 years or

so,

the efficiencies of TE ooling

systems remain low.

Mathiprakasam et al. (1991) studied TE cooling for MAC. Their findings

indicate that a

TE

MAC system could achieve a COP of only 0.43, which is much lower than those

of conventional M AC systems.

Lowi (1975) obtained a patent for an ejector MA C system powered by eng ine waste heat, and

h e discusses the basic principles

of

the system in his patent.

A

paper

by

Balasubramaniam

et

al.

(1976) extends Lowis study in indicating that more th an 70% of MAC energy consumption could be

saved by using this novel concept. A paper by

b w i

e t al. (1977) emphasized th e potential benefits

of a lightweight waste-heat-operated ejector MA C system. Ha mn er (1981) wrote a pa per discussing

the application of an ejector MAC system with detailed modeling of th e ejector. T h e advantages,

limitations, and recommendations for fu ture research an d dev elopm ent ar e given.

Akerman (1971) experim entally investigated th e suitability

of

absorption refrigeration cycles for

use in

MAC

systems.

He

found tha t the absorption cycle hea t rejection ra te

is

about

2.5

t o

3.5

times

as large as that of a vap or compression cycle. Ch arte rs and Meg ler (1974) studied th e feasibility of

an absorption MAC. The ir conclusion was that the main problem is selecting th e prop er refrigerant

and absorbent. Ballaney, Grov er, and Kapoo r (1977) considered absorption

MAC

feasible. Their

study indicates that lithium bromidehater absorption systems provide better results than

am m on ia ha te r systems. Da ta Analysis, Inc., (1980) developed a conceptual absorption MA C model

but failed to develop a prototype absorption

MAC

ystem. Mei, Chatu rvedi, and Lava n

(1979)

studied truck absorption refrigeration systems and concluded that absorption MAC is probably not

feasible for passenger cars because there would n ot

be

enough waste heat but th at it

is

feasible for

long-haul trucks. A concep tual system design is presented in their paper. Jackson (1987) did a

feasibility study

of

a vehicle w aste-heat-operated absorption system. He concluded that t he system

he analyzed was not feasible because

of

the high temperature

of

the engine waste heat, which

inhibited th e removal of hea t from th e evaporator. Schaetzle (1982) investigated t he solid desiccant

adsorption M A C system coupled with waste heat desiccant regeneration . Molecular sieve was

selected as th e desiccant material. A concep tual design and some preliminary tests

on

th e refrigerant

6


22/100


23/100

. .

.

. ..


24/100

3.

MACCOOLINGLOAD

Several studies of automobile cooling loads have appeared in the literature. Ru th

(1975)

experimentally validated a simple cooling load model which he derived. His model was used to

predict cooling loads for typical subcompact, compact, and standard cars a t various tem peratu res an d

relative humidities

(RHs).

Tables

2

and

3

show several

of

his

results.

Table 2 Predicted

automotive

cooling

load

requirements

City driving City driving Highway driving

Car

type

Ambient

(cool

down)

30

mph

60

mph

condition no outside air

100%

outside air

100%

outside air

( F/RH)

(Btu/h) (Btu/h) (Btu/h)

90/50

12,250 11,910

12,850

110/5% 13,170 12,950

13,940

Subcompact

100/20% 11,830 10,930

11,640

90/50%

14,140 14,220 14,120

Compact

100t20% 13,680 12,840 12,730

llO/5% 15,100 15,380 15,280

90/50% 17,270 17,620 17,520

Standard

100/20% 16,770 15,830 15,730

llO/ 5% 18,320 18,950 18,850

Table 3.

Breakdown

of

heat

load

Load Value (Btu/h)

of

Total Load

Solar

4470 34.8

Conductive

1770 13.6

Fresh Air

5400 42.0

Passenger

Instrument

lo00

200

7.8

1.6

TOTAL

12840 100.0

Ruth,

1975.

9


25/100

Ruth also applied his load model to the case where tinted windows are used to reduce solar

load, and

to the case of redu ced ventilation while the car is moving. His conclusions show that these

measures can significantly reduce the air conditioning cooling load.

Shimizu et al. (1982) have dev eloped a load model which is somewhat more d etailed than Ruth's

model. They consider th e

effects

that radiative heat transfer processes have on heat entering the

vehicle across th e roof, doors, and glass. T he surface radiation properties th at were

used

were

obtained directly from laboratory experiments. The y found that he at enterin g through th e roof can

account for 28% of the heat which enters a sedan that is parked in direct sunlight, the remaining 72%

being due to glass transmittance. Whe n th e vehicle is moving, th e forced ventilation load can account

for

51%

of th e cooling load, but with minimum outdo or air can be red uced t o 12%. Their transient

simulations showed that immediately after starting the car, the heat stored during hot soak can

account for 75% of th e instantaneous cooling load.

Sullivan and Selkowitz (1988) utilized a building thermal-load model called ESP which they

adapted for automo bile load analysis. The ir study concentrated o n th e relative effects of the three

radiative prope rties for glass: absorptance , reflectance, and transmittance, during various hot-soak

conditions. Th ey concluded tha t if

transmittance

is

reduced by increasing the reflectance, then internal temperatures can be greatly

reduced.

If

the transmittance is reduced by increasing absorptivity, then the glass will heat up and

in turn he at th e car, thus limiting the potential for reducing the hot-soak tem peratu re.

Their calculations did not extend

to

moving vehicles.

Dieckmann and Mallory (1990)

composed a numerical model which was very similar in scope

to

that

of

Shimizu et al. O n e significant difference was that their treatment

of

transient loads was

implicit - he user must prescribe the actual rate of cooling - hereas Shimizu et al. treated the

transient

case

explicitly by solving th e time-dep endent differential equations. Diec kmann and Mallory

used the model to simulate a number of load-reducing measures including wavelength-selective

glazing, roof and other insulation, ventilation during hot soak, and electrochromic glazing.

Finally, we n ote that the essential heat transfer processes that a re involved in determining th e

stationary hot soak temperature of an automob ile are relatively

few

in number. T h e following over-

simplified load model illustrates this by considering one representative car body, and by exploiting

some of the results of the above authors.

When the car is stationary and positioned in direct sunlight, Shimizu

et

al., among others have

found th at h eat e nte rs th e car by transmission through th e body glass, and by conduction through the

roof.

We denote these heating rates as

Q

and Qmf, respectively. T he ra te of heat leaving the

interior, Q

is

du e to conduction across the various body surfaces

except

for he roof. Und er steady

sta te hot-soak conditions,

Q

=

Q

+

Qe

Representative values for

Q

can be found using the sports model sedan configuration

considered by Sullivan and Selkowitz (1988). T he solar loads are calculated using standard formulas

(see

ASHRAE Fundamentals, 1989). Table 4 summarizes t he calculations ma de for Ju ne 21, at 30 N

latitude, at 12 o'clock noon.

10


26/100

Table

4.

Window arrangements''

Radiation Heat transfer

Area (ft2) Angle (deg) transmittance, (B tu h)

Front

12.3 31 265 3,260

Side

8.42 57 150 1,263

Total

10.613

Btu/(h-ft2)

Rear 20.3 19 300 6,090

Sullivan and Selkowitz,

1988.

T he he at transfer result, Q

is

based on a total glass transmittance r g= 0.83. For any other

r k Q can

be

calculated by the following correlation:

Q

=

10,613 (r J0.83 )

.

(2)

Regarding Qmf, the value calculated by Shimizu e t al. for a dark colored se dan un der similar

outdo or conditions may be used as a rep resentative value:

Q,=

1,706 BTU . (3)

T he heat loss rate Q may by calculated by using the correlation,

Q,=CAT,

(4)

where A T

is

the difference in temperature between ambient air and the car interior, and C is a

constant which depends only on th e car body configuration. For the sports model sedan

represented in Table 4, Sullivan and Selkowitz found that A T

= 32C.

Combining eqs.

1-4,

and

solving for

C,

we find that for this particular car,

C =

385

B T U h

-

C.

Now, or any rk the hot

soak tempe rature difference

A T

is given by

1,706 + 10,613 (rJ 0.830)

385

AT

=

Equation

( 5 )

predicts to within

8%

the

A T

calculated by Sullivan and Selkowitz for the range

0.23 < zg < 0.83 for the sports model sedan.

Finally we note that

in

Eq.

5)

tg an be replaced by an effective transmittance, tha t

is,

t he

fraction

of

incident radiation reaching t he car interior. This is useful in estimating the cooling effect

of window shades, for example.

11


27/100


28/100

Several alternative refrigerants have be en identified

as

potential R-12 replacements. Som e of

these are

R-134a fo r new cars;

ternar y blend R-22/R-152a/R-114 (or R-124) for retrofit;

oth er blends, such

as

R-22/R-l42b and R-22/R-124; and

non-inert refrigerants, such as R-152a and propanes.

4.1 R-134a

R-134a has many advantages as an R-12 replacement. Its thermodynamic properties are close

to those

of

R-12. Most importantly, it doe s not contain chlorine, so it will not deplete the ozone.

T he g lobal warming effect

of

R-134a is very low compared with that of R-12, and its toxicity (short-

term)

is

also low. However, som e changes are required when R-134a

is

used, such as changes in the

oil

and desiccant.

A

list

of

MAC

hanges for R-134a

follows:

A

parallel flow condenser is needed .

The thermal expansion valve needs to be changed.

An

oil

-

polyalkylene glycol

(PAG)

oil

- is

needed.

T he desiccant has to be ch anged from 4A-XH-5' to 4A-XH-7'.

Nylon lined hoses are needed.

R-134a has also been considered

as

the refrigerant for retrofitting purposes. While th e retrofit

could b e do ne technically, th e cost at this time could be prohibitively high. T he following items are

some

of

the major concerns in using R-134a to retrofit MAC.

The discharge pressure with R-134a is about 30-40 psi higher than with R-12 if cond enser and

expansion devices are not changed.

The refrigerant

cost is

high ($5.00 per lb).

Th e cost of changing refrigerant hoses alon e (if the old hoses are not nylon inner-lined ) will be

aroun d $200 to $300 (Dieckmann and Bentley).

T h e existing system must

be

thoroughly flushed and cleaned

to

avoid

the

decomposition

of

P A G

oil.

T h e desiccant must be changed a t a cost

of

around $80.

4.2

TERNARYBLENDS

Ternary blends a re m ixtures of thr ee refrigerants (R-22/R-152a/R-114 o r R-124) that have been

tested extensively as alternatives to R-12 (Bivens

et

al. 1989, Bateman et al.

1990).

Th e ternary

blends have t he following advantages:

*Product part number

of

UOP, Inc.

13


29/100


30/100

5.

NON-CFC

ALTERNATIVE

MAC

SYSTEMS

Th e alternative MAC systems ca n

be

divided in to work-actuated and therma l energy powered,

or heat-actuated, systems. T he work-actuated systems include conventional MAC units, air cycle

systems,

TE

cooling systems, and Stirling cycle systems. Heat-ac tuated systems include metal hydride

systems, desiccant MAC, ejector

MAC,

and absorption systems.

We

are well versed in som e of the

work-actuated MAC systems, such as the hermetic M AC system. Heat-ac tuated systems, however,

have an important feature in that it

is

possible to use automotive waste heat as the heat source to

power suc h systems, which could sav e up

to

70% of the energy consumed, according to som e analyses

(Lowi 1975). Most

of

these systems do not use CF C fluids, which m eans they ar e environmentally

acceptable. With the possible phaseou t

of

R-134a in th e future because of its global warming effect,

and with the potential of alternative systems for saving energy, R&D work on alternative MAC

methodologies is becoming very important. Th e following MA C opera ting conditions and com pressor

efficiency are assumed as a base for comparing different work -actuated systems:

Car interior temperature:

Am bien t air temperature: 100F.

Refrigerant exit temp. (condenser):

Refrigerant exit temp. (evaporator):

Com pressor efficiency (isentropic): 0.7.

Thermoelectric material properties:

77F and 60% RH

150F, subcooled to 140F.

4WF, superheated

to

50F.

Seeb eck coefficient: 1.8(2T+ 1985) lo-'

VPC;

electrical resistance: (6T

+

1735) 10 o h d c m ;

thermal conductivity: 0.0324 WlcmPC.

5.1

WORKACI'UATJZD MACSYSTEMS

5.1.1 Rehigerant

Vapor

Compression

Cycle:

Hermetic

Systems

Hermetically sealed

MAC

systems minimize refrigerant leakage. How ever, most previous studies

of this

type

of MAC ystems focused on high energy efficiency. Re cent efforts to improv e fuel

efficiency have resulted in sm aller automobile engines. T he air-conditioning load from a cycling fmed-

displacem ent com pressor can significantly affect vehicle drivability and perfo rma nce with small

engines. Variable-displacement compressor MAC ystems were develop ed to remedy this problem.

A paper by Bessler and F orbes (1987) discusses the application of dc motors to hermetically sealed

MAC systems. Th eir system uses a variable-speed brushless dc mo tor

to

drive a fmeddisplacement

compressor, thus achieving continuous control of cooling outpu t. Com pare d with variable-

displacem ent systems, th e electrically driven M AC provides th e following additional benefits:

A hermetic motor/compressor assembly means no shaft seal refrigerant leakage.

T he compressor is sma ller and less complex.

The packaging is more flexible (no drive belt).

Full cooling capacity can be achieved at any engine speed.

Conditioned air temperature can be controlled without re heat.

Herm etic M AC can easily be installed o n electric cars.

15


31/100

Two 5-hp brushless dc m otors were developed with maximum op erating spe eds

of

6ooo

and

7000

rpm. Th e researchers found that in order t o reduce th e system power consumption to a reasonable

level, fresh air intake had to

be

limited to aroun d

30%. A

2-hp motor was considered a reasonab le

cho ice with a cooling capacity

of

about

1.5

ton, which

is

more efficient than conventional MA C at

around 1.9 hphon . T he electrical M AC system needs at least 48

V of

dc to drive th e motor. Since

th e efficiency of th e alter nator is a direct multiplier o n th e electric MA C system efficiency, its design

is

critical, and it must

be

at least

75%

efficient to allow th e electric drive to match th e efficiency

of

conventional belt-drive systems. This study indicated that th e herm etic MA C system would require

a con trol strategy, compressor, and electrical system tha t a re different from th ose used in today's

automobiles.

Akaban e e t al.

(1989)

and Ikeda e t al. (1990) discuss a M AC system using a hermetically sealed

electrical air-conditioning system with a variable-speed scroll compressor coupled with a brushless dc

motor for electric vehicles.

Instead of using the engine to measure the power requirement, the

researchers designed a test stand to drive either an op en end compressor or an alternator. T he car

was tested in an environmental chamber.

Figure

1

shows the function blocks of the electric

compressor drive system. A high-performance cond enser with a

75%

higher coefficient of total heat

transfer was used t o replace th e original condenser. Th e test results showed that e ven with the hea t

flux

eduction and the change

of

condenser, th e total efficiency of

39.9%

was lower than tha t of the

baseline case (conventional system)

of 67%.

The electric MAC also has a COP

of 1.53,

which

is

lower than th e baseline COP of

1.81.

r -

I

I

Rotor

Paition

I

I

I

Signal

PoWa

(ac

or dc)

Fig. 1.

Electric

compressor drive

system.

16


32/100

Th es e studies indicate tha t limiting fresh air intake and reducing cooling load'(e.g., by means

of

window glazing) would b e necessary for electrically driven MAC. Substantial modifications o n th e

electrical system are also needed.

A recent newsletter by Nartron Corp.

(1991)

makes claims for

a

novel, herm etically sealed MAC

system with a n electrically driven turbine com pressor coupled with a Du Pon t low-pressure non-CF C

refrigerant. T h e system

is

compact and variable-speed. Th e newsletter further claims that th e system

has a

COP

bout 30% higher than that of conventional R-12 and R-134a MAC systems.

Based o n th e assumed operating conditions, the herm etic MAC

COPS

for R-134a an d R-22 can

be easily calculated a t around 2.25 an d 2.15, respectively. R-22 will have ab out 15% mo re cooling

capacity and will be op erate d close to 400 psia discharge pressure, versus R-134a op eratio n a t arou nd

278 psia.

5.12 Reversed BraytonAir

Cycle

Brayton air cycle air-conditioning systems ar e commonly used o n aircra ft because of their light

weight, compact size, and

the readily available bleed air. Figures 2a and 2b show schematics

of

a

reversed Brayton open- air cycle. T he air from the passenger compartment is sucked into a turbine

and then expanded, which lowers the air temperature. T he cold air exchanges hea t with ambient air,

and t h e cooled ambient air is then delivered to the passenger compartment. T he warmed air at low

pressure is compressed above ambient atmospheric pressure and vented. Th e turbocompressor would

operate at approximately

60 OOO

rpm, so it would be difficult to drive directly from th e automobile

engine. T he American Society

of

Heating, Refrigerating and Air-con ditioning Engineers (AS-)

Applications H andbook (1991) discusses the basic air cycle, the bootstrap cycle, and t he basic thre e-

whee l bo otstrap air cycle.

AMB IE NT

HEAT \ I A I R

EXCHANGER

I I 2

TURBI E

EXPANDER

COMPRESSOR

\ /

PASSE NGE R

COMPARTMENT

4

Fig.2a. Schematic

of Brayton

opencycle system.

17


33/100


34/100

e

e

e

Entropy

Fig.

26 Tem peratureentropy diagram of regenerative Brayton closedcycle system.

The results based on the calculations from Figs. 2c and 2d are shown in Table 7:

ambient tem perature:

100F;

recirculation temperature: 75F; and

supply air tempe rature: 50F.

Table

7.

Reversed Brayton cycle mobile air conditioning

calculated

performance

Case

1

Case 2

Turbine efficiency

0.9 0.9

Compressor efficiency

0.7 0.9

Cold-side heat exchanger temperature difference

F)

10.0

10.0

Hot-side heat exchanger temperature difference F)

10.0 10.0

Regenerator temperature difference

F)

10.0

10.0

Cooline COP

0.921 1.851

The computer code simulation indicated that the Brayton cycle is sensitive to the compressor

efficiency. An increase of com pressor efficiency from

0.7

to 0.9 will improve th e system cooling

COP

from 0.921 to 1.851.

19


35/100

5.13

Rotary-Vane

ompressor Air Cycle

Another type of air cycle MAC system is the rotary-vane air cycle system, called ROVAC in

many publications (Edwards

1975),

which uses a rotary-vane compressor

to

compress and expand air

simultaneously. Figure 3 shows the schematic of a ROVAC machine.

Air from the passenger

compartment is compressed and sent

to

the outdoor heat exchanger.

Air from the outdoor heat

exchanger

is

then expanded by the compressor.

N o

phase change

of

air occurs in eithe r the outdoor

hea t exchanger or the indoor heat exchanger. Because of th e isenthalpic effect during expansion, the

air becomes very cold before it is delivered to the passenger compartment. A report by the

engineering staff of Garret t (1977) indicates that ROVAC is not energy com petitive with oth er MAC

cooling methods. Tab le 8 shows the calculated and tested performance of t he ROVAC model

30B/45/9.5-4.

W A R M A I R I N

L

AIR

OUT

HEATEXCHANGER

INLEI :PORT

HEATEXCHANGER

OUTLETPORT

HEATEXCIANG3t

Fig.

3. ROVAC

mobile

air

conditioning

system.

20


36/100

Table

8.

Calculated and

actual

measured performance

data

of

ROVAC

model 30B/451954~

Computer predicted Actual measured

Param eter performance performance

Compressor outlet

Temperature

F)

316.0 315.0

Expander inlet

Pressure (psia)

47.75 48.2

Temperature

F)

105.5 108.0

Pressure (psia) 47.1 47.5

Expander outlet

Temperature

F) 9.0 34.0

Pressure (psia) 14.65 14.65

Rotor speed (rpm) 1510 1510

HP

drive 3.875 5.75

Air mass

flow

rate ( lbhr) 451.0 453.0

Cooling capacity (Bt uh )

29,979 23,284

C O P 3.04 1.59

Rotary-vane air cycle.

Garrett, Inc., Study of Reduction of Accessory Horsepower Requirement,

11th quarterly progress report

to

DOE, eport

74-310860

(33), 1977.

Th e m ajor differences between t he calculated and tested values are in th e air tem perature at

the exit of th e expander and th e power required to drive the unit. It looks as if a serious heat leak

is

in

the compressor,

yet

the calculated air mass flow rate is very close

to

the measured value, and

the calculated compressor outlet temperature is also very close

to

the measured temperature.

Without the ROVAC model or detailed description

of

the operating conditions

of

the machine,

further analysis of the system is difficult.

5.1.4

Thermoelectric cooling System

The

theory

of

TE

cooling is based on the Peltier Effect of certain materials. A

circuit

is formed

by two dissimilar materials, and

a

battery is introduced into the circuit to provide a direct current.

T he junction be tween the two dissimilar materials

is

heated or cooled. The heat evolved or absorbed

per unit time is proportional to the current flowing. TE modules could, therefore, use dc power

directly for cooling. This novel idea for MA C has several important ad vantages, such as no need for

refrigerant, adjustable cooling capacity, fast respons e, high initial cooling capacity, no moving parts

except a fluid circulating pump, and the ability

to

be

operated

as

a heat pump by reversing the dc

curre nt direction. TE ooling systems are a lso very rugged, which m eans little ma intenance isneeded.

Figure 4a shows the schematic of a

TE

MA C system and Fig.

4b

shows the schematic of a

TE

liquid-to-air hea t exchanger. Two

TE

heat exchangers are needed, on e acting as the condenser and

the o ther

as

the evaporator, with a circulating fluid to transfer heat from evaporator to condenser.

While on e

TE

heat exchanger could work theoretically, the efficiency would not be as high as it

would be w ith two

TE

heat exchangers. Air from the passenger compartment o r ambient air, at state

21


37/100

point

(SP) 1, flows

through the

TE

evaporator to be cooled and then in to the passenger

compartment. The TE ot side is then cooled by the circulating fluid, which is pumped to the oth er

TE

heat exchanger (condenser), whe re the circulating fluid is cooled. Ambient air at

SP

6 cools he

condenser TE hot side. A mathematical model was developed for this application (Mathiprak asam

et al.

1991,

see Appendix

B).

Including the input

of

realistic design factors and the shelf

TE

module

properties, Fig.

4c

and Table 9 show the calculated results of the system

COP as

a function

of

transfer fluid tem pera ture a t the evaporator inlet.

AMB IENT AIR

~r

3

Transfer

Fluid A L

CAR

INTERIOR

fig

4a.

Schematic of

thermoelectricmobile

air conditioning

system.

I I I I I I I I I I I I


38/100

4

.34

0.30

I

I

I

65

70 75

80 05

Tmnster FluidTempemre atCooler Inlet. T

I

Fig.

4c. &efficknt of performance of themmeleztric

mobile

air

conditioning

system.

Table

9. Thermoelectric

mobile air conditioning coefticient

of peromce

as

a function

of cooler inlet

heat

transfer

fluid

temperature

Heat transfer fluid

COP COP COP

Flow

rate Cooler inlet

Cooler

Rejector Overall

(gam) temp. F)

65

3.318 0.460 0.3197

400

70

2.363 0.546 0.3300

75

1.781 0.633 0.3303

80

1.389 0.721 0.3222

85

1.108

0.806

0.3064

65 3.836 0.516 0.36%

0.613 0.3830

0

2686

0.718 0.3872

5 2.009

80 1.564

0.833 0.3836

85 1.250

0.958 0.3732

500

65 4.234 0.548 0.4010

0.4171

00

70

9 5

0.653

0.4241

5 2.173 0.769

80 1.687

0.901

0.4236

85 1.348 1.050 0.4164

23


39/100

The maximum CO P the

TE

ystem can achieve

is

around

0.42

(parasitic power not included).

Realistic design factors

see

Appendix

B),

which include TE hermal resistance

of

off-the-shelf

E

modules, resulted in low COPs. For th e past

30

years or

so,

there has been no major breakthrough

in

TE

materials. Wh ile state-of-the-art

TE

modules are about

20%

more efficient tha n off-the-shelf

models, the current product has not reflected this technology. Even though th e TE M A C C O P is

low compared with that of conventional M AC systems, TE MAC'S unique advantages mentioned

previously sometimes ou tweigh its low efficiency.

A

passenger railway coach in Franc e ad opted a

TE

air-conditioning system (Stockholm e t al.

1982)

mainly because of its low maintenance requirement.

After a

3.5

year operation, there was no failure

of

the

TE

ystem, which confirmed its ruggedness.

TE

MAC could also

be

used on electric cars because

of

its simplicity.

If

TE

MAC systems we re

adopted by automobile manufacturers in large volumes, however, there could

be

a shortage of one

of the key elements, tellurium, unless a substitute material could be found.

5.15 Stirling Cyck cooling System

T he Stirling cycle theoretically could have high efficiency.

It

does not use CFC fluids, and it

can have modulated cooling capacity. A Stirling cycle heat pump was paten ted for automotive

heating and cooling applications; waste heat was considered

as

its power source (Kreger

1977).

A

study by ORNL on th e impact

of

CF C alternatives on energy (Fischer et al.

1991)

indicates that the

Stirling cycle for MAC could have a system

COP

of

1.7,

which

is

about

90%

that of current MAC

with a COP around

1.9. A

recent study showed that a kinematic Stirling cycle residential air-

conditioning system could have a COP of around

3.2

(Murphy 1991). Kinematic Stirling coolers can

be operated by carengine shaft power if

so

designed.

Stirling Therm al M otor

(STM)

as developed and demonstrated a variabledisplacement four

cylinder Stirling engine (God ett

1991),

which could lead to th e application of th e S tirling cycle to

MAC. Th e computer code performance projection of th e Stirling engine

MAC

COPs was between

1.6

and

2.0,

depending on the temperature difference between air temperature and Stirling heat

exchangers. Figure

5

shows th e schem atic of a Stirling cycle M AC system.

Comportment Ambient

toM

*r hot

-

r----y

_I ...

.-.....

Ambient

i

Compartment

ion

anger

Fs . Schematic

of

Stirling mobile

air

conditioningsystem

Table 10 displays the performance of the STM model, STM4-35

SAC,

y the computer

code

under the following assumed operating conditions:

24


40/100


41/100

Table

11

indicates that conventional

MAC

has th e highest

COPS.

Stirling cycle MAC seems to

have high efficiency, but the re

is no quantitative experimental data to back up th e calculated results.

Much

R&D

work

is

need ed for the above systems before they become practical.

53 HEATACI'UATJ3D MAC SYSTEM S

53.1 Ejector cool ing System

Figure 6a

shows

the schematic

of

an ejector cooling system, and Fig. 6b shows th e schematic

of

an ejector

MAC

system (Balasubramaniam e t al. 1976). T he system shown in Fig.

6b

can be operated

in thre e different modes: eng ine cooling, engine cooling and passenger c ompartm ent cooling, and

engine cooling and passenger compa rtment heating modes. T he theory of ejector

MAC

systems

is

similar to that of steam je t refrigeration, using high-pressure gas through an ejector

to

cre ate a low

gas pressure on a n evapo rator partially flooded with refrigerant. T h e low pressure

of

the refrigerant

causes low-temperature boiling, and the latent h eat

of

vaporization provides the cooling effect. T he

low-pressure refrigerant is compressed at th e divergent par t

of

the ejector wh ere refrigerant velocity

is reduced and pressure

is

increased. After condensing, par t

of

refrigerant

is

pumped

to

th e boiler

to produce high-pressure gas, and part

of

the gas

is

delivered

to

the evaporator.

COOLING

LOAD

WASTE

HEAT

Fig.6a chematic of ejector

cooling

system.

26


42/100

HIGHSDE

RECUPERATOR-

1

ENGINE

EXHAUST

I

MANIFoIl)

CONDENSER

ENGINE(300LING

JACKET

P U M P

-f-l

I

7

m

1

I

EXPANSION

VALVE

Fig.

6b. Schematic

of

ejector mobile air conditioning system

in

engine cooling

and airanditioning

m o d e

This is an a lternative with many attractive advantages, such as a minimum of moving parts, the

ability of the system to

be

actu ated with low-temperature waste hea t, high reliability, low ma intenance

cost, etc. Th er e have been several studies

of

ejector automotive MAC systems utilizing waste heat.

h w i (1975) had a paten t on an ejector M AC system. Balasubramaniam et al. (1976) studied the

energy impact of such M AC systems on cars. They concluded from their analysis that over 70% of

' fuel consumption used to run MAC could bessaved by using ejector MAC systems together with

reducing the system weight. Table 12 shows the calculated performance of an ejector M AC system

which indicates that th e CO P of the ejec tor system is only around 0.265 with R-11 as the refrigerant.

With an alternative refrigerant such asR-l34 a, the system CO P could be even lower. Ot her analyses

indicate that this type of system has a low COP at around 0.3 (Ch en 1978). Assuming that city

27


43/100

driving a t 30 mph consumes 1.5 gal/h

of

gasoline, that the heating value of gasoline

is

equal

to

Grade

1

fuel at 137,000 Btu/gal., that a car radiator dissipates 113 of total heat, and that 60% radiator heat

can be collected for ejector use, th ere will be only 41,000 Bt uh , which

is

on the borderline for ejector

M AC application with only

1

ton of cooling capacity. Because

of

the low system C OP , coupled with

better automobile fuel mileage in the future, passenger cars might not have enough waste heat

to

power an ejector MA C system. Appendix C presents an OR N L ejector model computer code for

performance calculation

of

the ideal case and, he case

of

constant-pressure mixing.

Table

12

Comparison

of

conventional mobile

air

conditioning

(MAC)

and

ejector mobile air conditioning

Waste-heatdriven

Conventional MA C eiector MA C

Cooling capacity (Btuh) 12,000 12,750

Refrigerant R-12 R-11

Energy input:

Shaft (hp)

Cooling jacket (B tu h)

Exhaust (Btuh)

Total (Btuh )

Energy rejected to ambient:

Refrigerant condenser (Btuh)

Engine radiator ( Btu h)

2.3

(compressor)

0.125

(purnpIb

-

4 4 , l W

-

4,000

6,OOo ,48,100

18,000 60,850

57,500 -

COP

2 . e 0.265

Evaporator:

Temperature

O F )

Pressure

Condenser:

Temperature O F )

Pressure (psia)

40

52

141

224

40

7

120

33

Refrigerant flow rate (lb/h) 190 560

Balasubramaniam et a1 (1976).

The

conventional'water pump is eliminated,

so

there

is

actually a negative shaft power increment.

'Assumed

60%

jacket heat utilization.

dBased on compressor hp only. Th e thermal efficiency of the engine is not taken into account.

532

Absorption

cooling System

Th e most comm on hea tdriv en air-conditioning devices are absorption units. Figure

7

s

a

schematic of an abs orption cooling system. Evapo rated refrigerant is absorbed by ano ther fluid, such

as

ammonia, at the evaporator, and this fluid then is absorbed by water at the absorber, creating a

low pressure. T he refrigerant-rich fluid is then pumped through heat exchanger to the regenerator,

where heat is added to sepa rate the refrigerant from the absorbent. High-pressure refrigerant goes

to the condenser, the n to t he expansion device, and finally to th e evaporator. T h e refrigerant-lean

fluid is routed to the absorber to absorb evaporated refrigerant.

28


44/100

1

i

H E A T

I N

' r -

V A L V E

XPANSION {

v o::':ER

H E A T

1 H E A T O U T

EVAPORATOR

Fs .

Schematic

of single-stage abrption

cooling

systems

The system needs only a fluid circulating pump to work. An absorption system rejects 2.5 to

3.5

times as much heat as a vapor compression system. Usually th e heat rejection

is

at a lower

tempera ture, which results in very large heat exchangers compared with those of conventional MAC

systems. Abso rption systems need

a

pair

of

fluids.

The most commonly used fluid pairs are

am mo niaha ter and waterbthium bromide. R-Wdimethyl ether of tetrae thylen e glycol (DME-TEG)

has also been considered. Th e estimated system cooling CO P will be less than 0.3 ( M e r m a n 1972).

Because

of

its low COP , absorption MA C

was

not considered feasible for passenger cars. Charters

and Megler 1974) have also studied an absorption cooling system for passenger cars. They

concluded that the main problem is the proper selection

of

refrigerant and absorben t. Mei,

Chaturvedi, and Lavan (1979) concluded that absorption refrigeration systems could be feasible for

long-haul trucks. When fully loaded, such trucks usually get

5-7

miles per gallon of diesel fuel.

Figure

8

shows that there

is

not enough waste heat from passenger cars to power such systems,

assuming

50%

exhaust gas hea t collection and 0.5 system COP. Figures 9 and 10 show the schematic

of a truck absorption refrigeration system. Assuming 30% of fuel energy to be exhausted at a

temperature around 7O0-8OO0F, the re will

be

enough energy to power a 3-ton refrigeration system.

One of these researcher's concerns is that the exha ust gas pressure dro p could possibly affect the

truck eng ine performance. Their ca lculation shows that with proper design of exhaust gas passage

through the generator, the pressure drop can be reduced to a negligible level.

29


45/100

o

10096 energy collection ?om exhaust gas

(ideal case)

50

energy

collection from exhaust gas (real case)

3 40 s o 60

10

2 0

speed,

milem

Fs .

Absorption

system cooling capacity vs

car

speed.

u

Exhaust Gas

t

F%

9. Schematk of truck

absorption

n&igemition system.

30


46/100

U u f f l c r

Truck

Radiator

Fs 10. Arrangement

of truck absorption

drigeration

systemcomponents

Phillips (1990) has analyzed many advanced absorption cycles at standard Air-c onditio ning and

Refrigeration Institute ARI) ated conditions (95F ambient for summer and

47F

for winter).

Table 13 shows the analytical results of some advanced absorption cycles for 80,000-Btu/h output.

The one with a generator-absorber heat exchanger (GAX) shows great promise for future

development

because

it

has

a structure of a single-stage machine yet has the capacity of a double

effect unit.

This

typeof system cou ld possibly

be used

for bus air-conditioning purposes w ith exhaust

gas as the power source.

533 Adsorption

Qde @esiocant)

cooling

System

Figure

11

shows the schematic of a desiccant cooling system pow ered by autom obile waste heat.

When w aste heat is applied to

a

desiccant bed,vapor refrigerant

is

regenerated, and a high pressure

is created. T he high-pressure refrigerant vapor

is

condensed into liquid at the condenser and goes

through an expansion device before being evaporated. Th e othe r desiccant

bed

adsorbs refrigerant

vapor and creates a low pressure. After th e cooling process is completed, the functions of the two

desiccant

beds

ar e reversed to start ano ther cooling process. In practical design, however, more tha n

two

desiccant

beds

could

be

needed to insure continuous supply

of

cooling capacity.

31


47/100

Table 13. Summary of advanced absorption cycle analyses

for

80,000-Btulh output

Variable

effect

1.04

2.04

35

Two-stage

GAX

1.06

2.06

45

7.5 39

6811351265

681265

No need Complex

~

Double

effectvcle

2R

COP cooling 1.11

0.83

I

0.79 1.03

1.03

COP heating 2.11 2.03

.83 1.79 2.03

388.4 110

heoretical

pumping

power (watt)

Pressure level

(psia)

NO. O f DUmDS

36

67128511250

151681265 681265 681265

681265

plus

6

intermediate

w

t

2

2 2

1 2

2, 1 multistage

8

7

1 4

8o. of major

components

Cutoff limit

(OF)

20

I s

0

I

-50,

I

lo

-15

20

Complex Complex

No

need I Automatic

onvertibility

Complex

a

Am on id wa te r properties uncertain at high pressures.


48/100


49/100

adsorption isotherms of typical desiccants (Collier

1991).

The figure indicates that -molecular sieve

(4A)

is close to the desired desiccant isotherm. Future desiccant MAC designs should have fast

absorption heat dissipation in order to achieve a better cooling effect and more frequent cycling for

higher cooling capacity.

fig. 1 Schematic of desiccant mobile air conditioning component arrangement

Inlet Refrigerant

Exhaust Gas Flow

Passage

-

wutlet Refrigerant

Fw 12b. Schematic

of desiccant bed desiga.

34


50/100

Fig.

13. Adsorption isotherms of typical desiccant mate rials

53.4 Metal

Hydride cooling

System

Hydriding alloys are intermetallic absorbent compounds that can absorb a very large quantity

of

hydrogen gas, and this process

is

reversible. Th e sorption and desorption processes ar e exothermic

and endothermic reactions, respectively. It is during the desorption process tha t th e cooling effect

is

achieved.

Figures 14a, 14b, and 14c present schematics of a metal hydra te cooling system. Initially it is

assumed that the temperature

of

the

MAC

system

is

ambient temperature,

T,

(point

2,

Fig.

14c).

When high-temperature waste heat is added to high-temperature hydride material, M,, t he

temperature and pressure increase to

T,

(point 3). At point 3, M, starts desorbing hydrogen. T he

low-tem perature hydride material,

MI,

at point 4 starts absorbing hydrogen gas (Fig. 14a). T he

absorbing heat is dissipated to ambient air.

M,

is then cooled down to am bient tempe rature and

coupled with the pressure drop back to point 2. Wh en the lower valve open s (Fig. 14b), MI starts

desorbing hydrogen gas becau se of th e low pressure, and its temperature drops to TL, hich is lower

than am bient temperature. T he cooling effect is achieved at point 1.

35


51/100

Fig.

1 4 a

Schematicof metal hydride cooling system, regeneration mode.

CLOSE

Fig.

14b.

Schematic

of

metal hydride cooling

system,

absorption m od e

ob

kat

input at TI

Q

reiedea

T

36


52/100

Me tal hydriding MA C systems can p otentially be operated with waste he at from engine exhaust

gas and can potentially have high COPS. They do not use CF C fluids, and they have a fast response

rate. A paper by Horowitz et al. (1979) presents a metal hydride heat pump application concep t by

using tubeswith high and low hydriding materials on each end of the tubes. It claims th at this system

could be used as a MA C system. Reilly and Sandrock (1980) mentions that Benz has dem onstrated

a bus running on hydrogen fuel with metal hydride to store hydrogen. M AC can

be

accomplished

simply by tak ing advantage

of

the low temperature

of

the hydriding materials when they are in the

desorbing mode. Figure 15 shows the schematic of a hydrogen fuel bus with metal hydride MAC.

ENGINE

WIAUSTGAS IN

/

/

HYDROGENCHARGING

PORT

W A R M A I R I N

Fig.

15.

Schematic

of

metal

hydride

mobile

air

conditioningsystem

for

hydrogen-fueled

vehicle.

The schematic shows that th e bus relies on thre e m etal hydride beds for hydrogen storage, two low-

temperature beds (beds 2 and 3, iron-titanium) and on e high-temperature bed (bed 1,magnesium-

nickel). Bed

1 is

heated by the exhaust, mainly steam, from the engine, which

is

also an auxiliary

heater for the bus.

Bed

2

is also heated by the engine exhaust, partially condensed steam from bed

1. Bed 3 encloses a liquid heat exchanger to provide air conditioning for the bus. Ron,Kleiner, and

37


53/100


54/100

Hvdride expansion.

Substantial volume changes are associated with hydriding and dehydriding

reactions. Typically, LaNi,, fo r example, expands by about

25%

during hydriding and co ntracts

an eq ual amou nt up o n dehydriding.

Metal hydride materials are not costly, and a system using them could have a fast response.

Most hydride materials are very brittle, however, which should also be considered in the system

design. It

is

estimated that a

1-

to

1.5-

cooling ton system requires

50

lb

of

hydriding materials.

Appendix

D

shows a sam ple calculation of a metal hydride cooling system with LaNi,&, and LaNi,

as

th e high- and low-temperature hydriding materials. Th e sample calculation indicates that a m etal

hydriding cooling system could have a first law

COP

(thermal) higher than

0.9.

How ever, this

COP

is

less than

19% of

its Carnot efficiency.

Metal hydride

MAC

is

definitely feasible for hydrogen fueled cars and buses. This

is

considered

a long-term option only because of the huge investment required for power plants, hydrogen

gene ration facilities, pipelines, and o the r necessary installations. Certa in shor t-term app lications, such

as

the operation of fleets of hydrogen powered vehicles serviced by central stations (Reilly and

Sandrock

(1980),

are n ot limited by t he above restrictions.

5.4

COMPARISON

OF

HEAT-ACI'UATED

MAC

SYSTEMS

Each

of

th e fou r systems selected in this study appe ars to have unique advantages and limits.

Excep t for absorp tion systems, oth ers a re not well studied. Experim ental data are, the refore , lacking.

Comparison

of

the criteria, such as the system cooling

COPS ,

could be misleading without

consideration of oth er design or operating factors. Th e comparison presented in Tab le

14 is

qualitative and should

be

considered fo r refer ence only.

Table 14.

Comparison

of

heat-actuated mobile air conditioning systems

M AC Estimated Advantages Limitations Rem arks

Ejector

0.3

Lightweight, compact, Low C O P Eng ine cooling

reliable capability

Absorption

0.79

- Ma tured technology Bulky and heavy Possible for truck

1-00 for conventional comp onents refrigeration and

machines

bus

M A C

system

COP

Adsorption

,

0.75

Low cost materials Adsorp tion heat Experim ental data

Metal

0.9

Fast response High thermal Possible for

hydride expansion, brittle, hydrogen fueled cars

dissipation problem based on

R-12

and heavy

'*Thermal

COPS

39


55/100

? , '


56/100


57/100


58/100

7. CONCLUSIONS

Almost all automotive manufacturers consider R-134a as the alternative refrigerant for

MAC

application. Phase-in has been planned for between 1991and 1995. However, major uncertainty has

been associated with possible future regulatory action against R-134a for its GWP. I t is therefore

prudent

to

examine the non-CF C alternative cooling technologies. This report studied two types

of

MAG,

work-actuated and heat-actuated.

For work-actuated

MACs,

some of the technologies are familiar, such as the hermetic system

and Brayton air cycle MA&. Current hermetic systems almost exclusively use

CFC-,

hydrochlorofluorocarbon- (HCFC), or hydrofluorocarbon- (HFC) type fluids. If a non-CFC type

refrigerant is developed for MAC application, hermetic MAC could be a viable alternative

technology. Brayton cycle systems are desirable because air o r othe r inert gases can b e used as

working fluids, but th e system

COPs

need

to

be further improved before these systems are practical,

The Stirling cycle MAC is very attractive for its high theoretical COPs and potential long life.

Further development, however, is needed to experimentally prove that it is a viable MAC system.

Solid-state cooling TE MACs, even with their many advantages, will probably remain for special

applications if n o major

TE

material breakthrough

is

achieved.

Fo r he at-actuated MA C systems, the energy saving potential is high because

of

the possible use

of waste heat. However, most of them have eith er low system COPs or bulky components. Ejector

MAC

is

reliable and compact, but the calculated system

COPs,

with CFC fluids

as

the working

medium, a re only in the orde r of

0.3.

Further COP reduction is expected if non -CF C fluids are used.

With t he automobile engines becoming more efficient in t he future, it is possible that the re would

not b e enoug h waste heat for ejector cooling. Absorption machines a re a m atured technology with

bulky and heavy components that are not suitable for MAC applications. How ever, for long-haul

trucks with an enormous amount of waste heat at a tem perature around 700 to 8oo F, absorption

systems might be feasible for low-temperature truck refrigeration application for perishable foods.

Adsorption, o r desiccant, MACs could possibly use water

as

th e working medium. Fas t dissipation

of adsorption and desiccant rege neration he at for efficient system operation could be a problem th at

would lead to large hea t exchangers. Me tal hydride MAC systems have a fast respo nse and high

theoretical COPs, and they need a set of hydriding materials to function. Since hydriding materials

ar e intermetallic compounds, they ar e quite heavy, and that

is

not desirable for M AC application.

However, if hydrogen is used as the automotive fuel, hydriding materials could be used as the

hydrogen storage mediums. Me tal hydride MAC omes

with

hydrogen-fueled cars because the

relea se of hydrogen from hydriding materials will cool th e metal hydride

beds

that can be used for

M A C application.

The

tudy indicates that each alternative MA C technology has its merits and limitations. Mo st

of them req uire further

R&D

work. With R-134a facing possible futu re regulatory action and with

no suitable non-CFC replaceme nt refrigerant in sight, R & D

of

selected alternative M AC technologies

has become urgent and important because of the long lead time required for such systems to be

practical.

43


59/100

I


60/100

&REFERENCES

Akabane, H., et al., Evaluation

of

an Electrically Driven A utomo tive Air C onditioning System Using

a Scroll Hermetic C ompressor with a B rushless D C M otor, S A E Paper 890308, presented at the

Society of Automotive Engineering International Congress and E xposition, Detroit, February

27-March

3,

1989.

M e r m a n , J. R., Autom otive Air Cond itioning Systems with Absorption Refrigeration, S A E Paper

710037, Society of Automotive Engineers, Warrendale, PA, 1972.

Ally, M. R., W. J. Rebello, and M. J. Rosso , Jr., Metal Hydride Chemical He at Pum p or Industrial

Use, pp. 686-93 in Symposium of the Sixth Annual Industrial Energy Conservation Technology

Conference

&

Exhibition,

Vol. 2, Intersociety Energy Conversion Engineering Conference 1984,

Houston , April 15-18, 1984.

A R I R

h

TReferenceList,

Research and Technology Departme nt, Air-conditioning and R efrigeration

Institute, Arlington, VA, January 1991.

ASHRAE W AC Ap plic ation Handbook,

Ch. 9, American So ciety of Heating, Re frigerating, and Air-

Conditioning Enginee rs, Inc., A tlanta, 1991.

Balasubramaniam, M., et al., Fuel Economy Potential of a Combined Engine Cooling and Waste

H ea t Driven Automotive Air-Conditioning System, pp. 25-32 in

11th Intersociety Energy

Conversion Engineering Conference, State Line, NV,

Alternative Non - CFC Mobile Air Conditioner

Documents