Technical Paper

IIAR Natural Refrigeration Conference

& Heavy Equipment Expo

March 24 – 27, 2024

Orlando, Florida

ACKNOWLEDGEMENT

The success of the IIAR Natural Refrigeration Conference & Heavy Equipment Expo is due to
the quality of the technical papers in this volume and the labor of its authors. IIAR expresses
its deep appreciation to the authors, reviewers, and editors for their contributions to the
ammonia refrigeration industry.

ABOUT THIS VOLUME

The views expressed in this technical paper are those of the authors, not the International
Institute of Ammonia Refrigeration. They are not official positions of the Institute and are not
officially endorsed.

International Institute of Ammonia Refrigeration

1001 North Fairfax Street, Suite 503
Alexandria, VA 22314

703-312-4200 • info@[Link] • [Link]

© 2024 IIAR
 Technical Paper #2

CO2 Systems Add-Ons:

Calculations and Field Measurements

Giacomo Pisano, CO2 Business Development Manager,

and Nabih Hussein, Refrigeration Sales Engineer
DORIN USA

Abstract

This paper describes enhanced and integrated carbon dioxide refrigeration systems that can improve
yearly energy figures when compared to traditional flash-gas bypass (FGB) solutions, making CO2 an
attractive option for industrial refrigeration in some regions.

The typical FGB solution represents the simplest transcritical carbon dioxide (CO2) system configuration,
but it suffers energy efficiency penalties with higher heat-sink temperatures, leading to unfavorable
yearly energy efficiencies at warmer latitudes.

This paper provides a detailed description of various CO2 systems. Starting from FGB configuration,
in-depth energy efficiency analysis is provided, including job site field experience. Then, several FGB
add-ons are described, including parallel compression; liquid and vapor ejectors; adiabatic cooling; and
line start permanent magnet electric motors. Each add-on contribution is analyzed with calculations
and field measurements.

We also describe CO2 system integration, showing how heat recovery and air-conditioning can be
integrated into the same refrigeration unit, thus making any specific building completely self-sufficient
from an energy standpoint. The same CO2 system is able to perform both low- temperature and
medium-temperature duties, together with providing sanitary hot water, comfort cooling during
summer, and heating during winter. Again, this integration is analyzed from a theoretical standpoint,
and then concrete examples with field measurements are shown.

© IIAR 2024 1
 CO2 Systems Add-Ons: Calculations and Field Measurements

Introduction

Threats of ozone depletion and global warming have accelerated the use of carbon
dioxide (CO2) as a refrigerant in the last two decades, leading to development of the
flash-gas bypass (FGB) configuration, the most cost-effective execution for a CO2
transcritical system.

Heat sink temperature plays a key role for this system’s performance. In the case of
an air-cooled cycle, when ambient temperature exceeds 73°F, no condensation can
occur and the high-pressure heat exchanger will not condense the CO2 stream, the
reason for it being called “gas cooler.” The heat-sink temperature determines the gas
cooler’s high pressure, which optimizes the system coefficient of performance (COP),
“gas cooler.” The heat-sink temperature determines the gas cooler’s high pressure, which
as shown in Figure 1.
optimizes the system coefficient of performance (COP), as shown in Figure 1.

Figure
Figure 1. 1. Determination
Determination of optimal
of optimal high pressure
high pressure withheat-sink
with a given a given temperature.
heat-sink temperature.

Figure 1 shows an air-cooled compression process starting from 430 psi (corresponding to 22°F
SST) with an ambient temperature of 90°F and assumes an infinite gas cooler surface equalizing
gas cooler outlet and heat-sink temperatures. With no condensation, infinite head pressures
fulfil the gas cooler outlet temperature constraint, and one may choose to end compressing
anywhere above#2
Technical Paper the critical pressure. This results
© in significant
IIAR 2024 performance differences with 3
varying discharge pressures. Figure 1 also shows that there is a specific end-of-compression
pressure optimizing the cycle COP, presented graphically by drawing the tangent line starting
from the critical point (87°F at 1045 psig) to the gas cooler outlet isotherm [1].
 2024 Natural Refrigeration Conference & Heavy Equipment Expo

Figure 1 shows an air-cooled compression process starting from 430 psi

(corresponding to 22°F SST) with an ambient temperature of 90°F and assumes an
infinite gas cooler surface equalizing gas cooler outlet and heat-sink temperatures.
With no condensation, infinite head pressures fulfil the gas cooler outlet temperature
constraint, and one may choose to end compressing anywhere above the critical
pressure. This results in significant performance differences with varying discharge
pressures. Figure 1 also shows that there is a specific end-of-compression pressure
optimizing the cycle COP, presented graphically by drawing the tangent line starting
from the critical point (87°F at 1045 psig) to the gas cooler outlet isotherm [1].

This optimal head pressure varies with heat-sink temperature: a lower higher-heat-
sink temperature will lead to a higher optimal pressure, as shown in Figure 2.

Figure [Link]
Figure Determinationof optimal high high
of optimal pressure with varying
pressure heat-sinkheat-sink
with varying temperature.
temperature.

Figure 2 also shows that a severe deterioration in performance takes place with increasing
heat-sink temperatures, with COP(90 °F) > COP(110 °F) where:

COP(90 °F) is the system COP with 90°F ambient temperature.

COP(110 °F) is the system COP with 110°F ambient temperature.

4 © IIAR
At the same time, a decrease in heat-sink 2024
temperature leads to a COP improvement,Technical Paper #2
especially
when it comes to subcritical system operation which will typically be in place for ambient
temperatures below 73 °F. Figure 3 shows a subcritical CO2 cycle which could be representative
of a mild day with 40°F ambient temperature and demonstrates a tremendous COP
CO2 Systems Add-Ons: Calculations and Field Measurements

Figure 2 also shows that a severe deterioration in performance takes place with
increasing heat-sink temperatures, with COP(90 °F) > COP(110 °F) where:

COP(90 °F) is the system COP with 90°F ambient temperature.

COP(110 °F) is the system COP with 110°F ambient temperature.

At the same time, a decrease in heat-sink temperature leads to a COP improvement,

especially when it comes to subcritical system operation which will typically be
in place for ambient temperatures below 73 °F. Figure 3 shows a subcritical CO2
cycle which could be representative of a mild day with 40°F ambient temperature
and demonstrates a tremendous COP improvement, with COP(40 °F) >> COP(90 °F).
Calculations show that a 4% increase in COP can be gained for each 2°F in ambient
temperature reduction during transcritical operation. When subcritical operation is
in place, COP improvement is even more consistent. For instance, going from 90°F
ambient to 40°F leads to a 300% COP improvement.

Figure 3. CO2 system cycle with 40°F and 90°F heat-sink temperatures.
Figure 3. CO2 system cycle with 40°F and 90°F heat-sink temperatures.
It is also interesting to note how CO2 allows operation with much lower condensing
temperatures when compared to standard hydrofluorocarbon (HFC) cycles: in fact, the higher
operating pressures always provide enough pressure differential for the metering device to
function correctly, permitting condensing temperatures as low as 40°F, when a consistent gas
cooler design is in place.
Technical Paper #2 © IIAR 2024 5

Considering these arguments, it is clear how standard FGB CO2 systems can offer great yearly
efficiency levels in cold-to-mild environments, while efficiency penalties must be accounted for
 2024 Natural Refrigeration Conference & Heavy Equipment Expo

It is also interesting to note how CO2 allows operation with much lower condensing
temperatures when compared to standard hydrofluorocarbon (HFC) cycles: in fact,
the higher operating pressures always provide enough pressure differential for the
metering device to function correctly, permitting condensing temperatures as low as
40°F, when a consistent gas cooler design is in place.

Considering these arguments, it is clear how standard FGB CO2 systems can offer
great yearly efficiency levels in cold-to-mild environments, while efficiency penalties
must be accounted for in warmer environments. For this reason, over the course of
the last decade, an efficiency-restricted “CO2 Equator” has been defined in literature,
showing the territorial limitation for efficient FGB CO2 system execution when
compared to other refrigerants, depending on average ambient temperature (Figure
4). These limitations see annual energy consumption of a CO2 system larger than
an equivalent HFC system for average ambient temperatures above 59°F (15°C)
[2]. Therefore, in North American jurisdictions, a standard CO2 FGB cycle is only an
efficient solution above the blue line depicted in Figure 5.

Figure 4. Global annual average temperatures.

Figure 4. Global annual average temperatures.

6 © IIAR 2024 Technical Paper #2

 CO2 Systems Add-Ons: Calculations and Field Measurements

Figure 4. Global annual average temperatures.

Figure 5. Annual
Figure average
5. Annual temperatures
average and CO
temperatures 2 equator
and in North
CO2 equator in America.
North America.

CO2 System Add-ons

During the last decade, numerous CO2 system add-ons have been successfully
introduced in order to improve CO2 system efficiency in warmer environments (i.e.,
below the aforementioned blue line). This paper will address the following add-ons:
parallel compression, vapor and liquid ejectors, adiabatic cooling, and compressors
with permanent magnet electric motors.

Technical Paper #2 © IIAR 2024 7

 CO2 System Add-ons
During the last decade, numerous CO2 system add-ons
2024 Natural have been
Refrigeration successfully
Conference introduced
& Heavy Equipmentin
Expo
order to improve CO2 system efficiency in warmer environments (i.e., below the
aforementioned blue line). This paper will address the following add-ons: parallel compression,
vapor and liquid ejectors, adiabatic cooling, and compressors with permanent magnet electric
motors.
Parallel Compression

Parallel Compression
A simple FGB system can be modified by implementing parallel compression. In a
A simple FGB system can be modified by implementing parallel compression. In a typical
typical flash-gas
flash-gas bypass
bypass system, system,
the the
flash gas flash gas
building building
up in up in thepressure
the intermediate intermediate pressure
receiver is
diverted
receivertoward the compressors
is diverted toward thesuction, as shown
compressors in Figure
suction, 6.
as shown in Figure 6.

Figure
[Link]-gas bypass
Flash-gas system.
bypass system.

This bypass is in place because flash gas is not useful to heat exchange in the evaporators and it
isThis bypass
therefore is efficient
not in placetobecause
circulateflash gas is
this back not
and useful
forth fromtoand
heat
to exchange in theespecially
the evaporators,
inevaporators and itdistributed
the case of large is therefore not efficient to circulate this back and forth from and
systems.
to the evaporators, especially in the case of large distributed systems.
Instead of bypassing intermediate receiver flash gas, parallel compression is a kind of execution
where an additional set of compressors is in place, sucking gas directly from the intermediate
Instead receiver,
pressure of bypassing intermediate
as shown in Figure [Link] flash gas, parallel compression is a kind
of execution where an additional set of compressors is in place, sucking gas directly
from the intermediate pressure receiver, as shown in Figure 7.

8 © IIAR 2024 Technical Paper #2

 CO2 Systems Add-Ons: Calculations and Field Measurements

[Link]
Figure Parallel compression
compressionexecution.
execution.

During warmer seasons, a significant portion of the total CO2 mass flow exiting the
During warmer
high-pressure valveseasons, a significant
consists of portionupofpressure
flash gas, building the totalinCO mass flow exiting
the2 intermediate the
pressure
high-pressure
receiver, valve
accounting forconsists
as much of
as flash
45% ofgas,
thebuilding
total massupflow
pressure
leavinginthe
thehigh-pressure
intermediatevalve.
Bypressure
using parallel compression,
receiver, accountingthisfor
large amountasof45%
as much refrigerant
of the mass can beflow
total mass compressed
leaving from
the
the intermediate receiver pressure to the gas cooler pressure, instead to the
high-pressure valve. By using parallel compression, this large amount of refrigerant evaporator
pressure, leading to a COP improvement in the range of 7% at peak temperature [3].
mass can be compressed from the intermediate receiver pressure to the gas cooler
pressure,
Liquid andinstead to the evaporator pressure, leading to a COP improvement in the
Vapor Ejectors
range of
Ejectors are7% at peak
a means temperature
to further improve [3].
CO2 system efficiency. They convert high-pressure
potential energy in the motive flow (gas cooler outlet refrigerant stream) into kinetic energy,
drawing a flow from a suction port (evaporator outlet refrigerant stream). Two main kinds of
Liquidare
ejectors and Vapor Ejectors
normally considered: liquid ejectors and vapor ejectors.

Liquid Ejectors
Ejectors are a means to further improve CO2 system efficiency. They convert high-
In typical direct expansion (DX) systems, a consistent part of the evaporator line is used in order
pressure potential energy in the motive flow (gas cooler outlet refrigerant stream)
to assure enough superheat and avoid liquid refrigerant returning to the compressors suction
intoThis
line. kinetic
leadsenergy, drawing a flow
to safe compressor from but
operation a suction portefficiency
introduces (evaporator outlet
losses sincerefrigerant
the
stream). Two
super-heating maininto
phase kinds
the of ejectors coil
evaporator are will
normally
featureconsidered: liquid
a very limited heatejectors
transfer. and
vapor ejectors.
The use of liquid ejectors allows the system to work with ideally no superheat at the evaporator
outlet, thus maximizing the heat transfer. The evaporator’s outlet is wet and streams into a
generously sized low-pressure receiver (LPR), where the liquid excess accumulates at the
bottom, to be sucked by the liquid ejector and compressed into the intermediate pressure
receiver. This process lifts the saturated evaporating
Technical Paper #2
pressure/temperature up to 9R, greatly
© IIAR 2024 9
improving system efficiency.
 2024 Natural Refrigeration Conference & Heavy Equipment Expo

Liquid Ejectors

In typical direct expansion (DX) systems, a consistent part of the evaporator line is
used in order to assure enough superheat and avoid liquid refrigerant returning to
the compressors suction line. This leads to safe compressor operation but introduces
efficiency losses since the super-heating phase into the evaporator coil will feature a
very limited heat transfer.

The use of liquid ejectors allows the system to work with ideally no superheat at the
evaporator outlet, thus maximizing the heat transfer. The evaporator’s outlet is wet
and streams into a generously sized low-pressure receiver (LPR), where the liquid
excess accumulates at the bottom, to be sucked by the liquid ejector and compressed
into the intermediate pressure receiver. This process lifts the saturated evaporating
pressure/temperature up to 9R, greatly improving system efficiency.

Vapor Ejectors

The flash gas amount accumulating at the bottom of the low-pressure receiver can
also be compressed by means of the vapor ejector into the intermediate pressure
receiver. This will lead to a significant increase in the mass flow available for
parallel compression, bringing further efficiency improvement. It also creates a need
for multiple parallel compressors to cope with variable load conditions. It is not
necessary for liquid and vapor ejectors to be implemented at the same time; however,
the inclusion of a liquid ejector will increase vapor-ejector effectiveness. The vapor-
ejector entrainment ratio (the amount of fluid the ejector is able to suck at given
boundary conditions) will increase with decreasing pressure differences; if a liquid
ejector is not in place, the vapor ejector will work with a larger pressure difference
due to the evaporation pressure lift introduced by the liquid ejector itself.

10 © IIAR 2024 Technical Paper #2

compressed by means of the vapor ejector into the intermediate pressure receiver. This will
lead to a significant increase in the mass flow available for parallel compression, bringing
further efficiency improvement. It also creates a need for multiple parallel compressors to cope
with variable load conditions. It is not necessary for liquid and vapor ejectors to be
implemented at the same time; however, the inclusion of a liquid ejector will increase
CO2 Systems Add-Ons:
vapor-ejector Calculations
effectiveness. The and Field Measurements
vapor-ejector entrainment ratio (the amount of fluid the
ejector is able to suck at given boundary conditions) will increase with decreasing pressure
differences; if a liquid ejector is not in place, the vapor ejector will work with a larger pressure
difference due to the evaporation pressure lift introduced by the liquid ejector itself.
The implementation of both liquid and vapor ejectors leads to further COP
The implementation
improvements of both
in the range liquid and vapor
of 16% ejectors
at peak leads
ambient to further COP
temperatures [3].improvements in the
Figure 8 shows
range of 16% at peak ambient temperatures [3]. Figure 8 shows a system layout featuring both
a system layout featuring both liquid and vapor ejectors.
liquid and vapor ejectors.

Figure
Figure [Link]
Systemwith
withboth
both liquid
liquid and
andvapor
vaporejectors.
ejectors.

Adiabatic Cooling
Adiabatic Cooling
An adiabatic gas cooler is made of water-wetted pads located outside the condenser coils; this
helps to decrease the CO2 gas cooler exhaust temperature and therefore delays transcritical
An adiabatic
operation gas cooler
of a booster is made
system. of water-wetted
A water pads
valve is triggered located
when outsidetemperature
the ambient the condenser
exceeds 70°[Link]
coils; this Watertoisdecrease
sprayed on
thethe
COpads, the gas cooler fans pull air through the wet pads,
2 gas cooler exhaust temperature and therefore

delays transcritical operation of a booster system. A water valve is triggered when the
ambient temperature exceeds 70°F. Water is sprayed on the pads, the gas cooler fans
pull air through the wet pads, and the air becomes humid and cooler prior to passing
through the coils. This significantly helps in making the system up to 16% more

Technical Paper #2 © IIAR 2024 11

 2024 Natural Refrigeration Conference & Heavy Equipment Expo

energy-efficient, depending on the specific job site location and ambient temperature
profile.

and the air becomes humid and cooler prior to passing through the coils. This significantly helps
in Permanent MagnetupMotors
making the system to 16% more energy-efficient, depending on the specific job site
location and ambient temperature profile.
Typical permanent magnets motors (PMM) are designed with magnets embedded into
Permanent Magnet Motors
the motor rotor, in most cases their stators feature the typical squirrel cage design,
Typical permanent magnets motors (PMM) are designed with magnets embedded into the
very rotor,
motor similar to that
in most of atheir
cases standard
statorsasynchronous motor.
feature the typical squirrel cage design, very similar to
that of a standard asynchronous motor.
PMMs are synchronous, this meaning there is no slippage between rotor and stator
PMMs are synchronous, this meaning there is no slippage between rotor and stator magnetic
magnetic field, leading to much lower electric losses.
field, leading to much lower electric losses.

Typical
Typical PMMs
PMMs needneed an external
an external drivedrive (inverter);
(inverter); however,
however, a new technology—line
a new technology—line start
permanent magnets (LSPM)—is now available from various electric motor
start permanent magnets (LSPM)—is now available from various electric motor suppliers. LSPM
electric motors
suppliers. are aelectric
LSPM hybrid configuration
motors are abetween the standard asynchronous
hybrid configuration between the design
standardand the
permanent magnets technology, as shown in Figure 9.
asynchronous design and the permanent magnets technology, as shown in Figure 9.

Figure 9.
Figure 9. LSPM
LSPMrotor
rotordesign.
design.

With this design it is possible to energize the electric motor directly online, with no need for an
With this
inverter. The design it isstart
motor will possible to energize
as a standard the electric
induction motor motor directly
and switch online, with no
to synchronous
need forwithin
operation an inverter. The motor will start as a standard induction motor and switch to
milliseconds.
synchronous operation within milliseconds.
Besides the cost effectiveness (because there is no need for an external inverter), this electric
motor configuration allows the ability to achieve a very interesting efficiency improvement. Lab
tests have demonstrated that compressor energy efficiency ratio (EER) improvements are above
12%, depending on the operating conditions, as will be shown in a later section.

12 © IIAR 2024 Technical Paper #2

Calculated Performance Improvements for CO2 System Add-ons
CO2 Systems Add-Ons: Calculations and Field Measurements

Besides the cost effectiveness (because there is no need for an external inverter), this
electric motor configuration allows the ability to achieve a very interesting efficiency
improvement. Lab tests have demonstrated that compressor energy efficiency ratio
(EER) improvements are above 12%, depending on the operating conditions, as will
be shown in a later section.

Calculated Performance Improvements for CO2 System Add-ons

The various CO2 system add-ons bring consistent efficiency improvements, which
are documented by means of a specific software [4], capable of simulating various
system designs, including flash-gas bypass, parallel compression, liquid and vapor
ejectors, adiabatic cooling, and permanent magnet motors. Calculations have been
made assuming the following system specifications:
• LT load: 3.000 kbtu/h @ -22°F SST
• MT load: 6.000 kbtu/h @ +22°F SST
• Max ambient temperature: 95°F

Calculations with a simple flash-gas bypass execution suggest a peak temperature

EER of 5.89. The same system designed with parallel compression execution, brings
EER to 6.30. Implementing both liquid and vapor ejectors further improves the EER
to 6.93. When adiabatic cooling is also considered, then EER is brought up to 8.03. If
permanent magnet motors are used for the medium-temperature suction group and
for the parallel compression group, then EER reaches the remarkable level of 8.75.
Figure 10 shows the various system configurations and the specific software output.

Technical Paper #2 © IIAR 2024 13

 Calculations with a simple flash-gas bypass execution suggest a peak temperature EER of 5.89.
The same system designed with parallel compression execution, brings EER to 6.30.
Implementing both liquid and vapor ejectors further improves the EER to 6.93. When adiabatic
cooling is also considered, then EER is brought up to 8.03. If permanent magnet motors are
2024 Natural Refrigeration Conference & Heavy Equipment Expo
used for the medium-temperature suction group and for the parallel compression group, then
EER reaches the remarkable level of 8.75. Figure 10 shows the various system configurations
and the specific software output.

Figure 10. Performance calculations at peak temperature for various systems.

Figure 10. Performance calculations at peak temperature for various systems.

Theoretical calculations show that the aforementioned CO2 systems add-ons are
confirmed to bring significant energy efficiency improvements; keeping the simple
flash-gas bypass execution as the baseline, calculations show that following
improvements are achieved:
• Parallel compression: +7%
• Ejectors: +11%
• Adiabatic cooling: +16%
• Permanent magnet motors: +9%

14 © IIAR 2024 Technical Paper #2

CO2 Systems Add-Ons: Calculations and Field Measurements

Since these results are derived through mathematical system modelling, it may be
argued they do not really represent field figures. The following section shows real-world
examples where field measurements were noted for several different CO2 systems.

Field Measurements for CO2 System Add-ons

Any refrigeration system in real-world operation is subject to boundary conditions,

which impact its efficiency. It is therefore difficult to calculate rigorously the
individual influence of any one of the CO2 systems add-ons previously described.
Nevertheless, some of these have been documented and recorded.

Adiabatic Cooling

A CSF supermarket in Beijing, China [5], has been equipped with a CO2 transcritical
refrigeration system. Especially during summer, the local weather is very dry, which
led engineers to choose an adiabatic gas cooler design. However, due to a minor
issue, the adiabatic gas cooler water pads were not put in operation during the first
days of the system operation, which was in July. Water pads were applied shortly
after this installation, with notable energy efficiency benefits.

During these days the main system operating parameters were monitored, with a
special focus on the gas cooler operation, with measurements shown in Figure 11.

Technical Paper #2 © IIAR 2024 15

 system. Especially during summer, the local weather is very dry, which led engineers to choose
an adiabatic gas cooler design. However, due to a minor issue, the adiabatic gas cooler water
pads were not put in operation during the first days of the system operation, which was in July.
Water pads were applied shortly after this installation, with notable energy efficiency benefits.

During these days the main system operating parameters

2024 Natural were monitored,
Refrigeration Conference &with
Heavya Equipment
special focus
Expo
on the gas cooler operation, with measurements shown in Figure 11.

Figure 11. Field measurements with and without adiabatic gas cooler operation.
Figure 11. Field measurements with and without adiabatic gas cooler operation.

During operation without water pads, and with an ambient temperature of 103°F, gas
cooler operating pressure was set to 1363 psi with a 76% HP valve closing and a gas
cooler outlet temperature of 108°F.

When water pads were added to the system, with a very similar ambient temperature
of 100°F, it was possible to significantly decrease gas cooler pressure to 1167 psi,
obtaining a gas cooler outlet temperature of 84.2°F: this shows how beneficial the
use of an adiabatic gas cooler is from an efficiency perspective, especially in very
dry areas, and confirms the prospected 16% EER improvements mentioned in the
previous section.

Liquid Ejectors

A Migros supermarket in Bulle, Switzerland [6], was one of the first sites where
ejector technology was deployed. At the time, some of the vending area evaporators
had been equipped with instrumented coils; the given cabinets were meant to keep
air outlet temperature at 31°F and thermocouples were attached to various coils
bends, so to monitor fluid temperature during system operation. The system was
made to run with and without liquid ejectors, with results being shown in Figure 12.

16 © IIAR 2024 Technical Paper #2

Liquid Ejectors
A Migros supermarket in Bulle, Switzerland [6], was one of the first sites where ejector
technology was deployed. At the time, some of the vending area evaporators had been
equipped with instrumented coils; the given cabinets were meant to keep air outlet
CO2 Systems Add-Ons: Calculations and Field Measurements
temperature at 31°F and thermocouples were attached to various coils bends, so to monitor
fluid temperature during system operation. The system was made to run with and without
liquid ejectors, with results being shown in Figure 12.

Figure 12. Field measurements for an adiabatic gas cooler operation with and without ejectors.
Figure 12. Field measurements for an adiabatic gas cooler operation with and without ejectors.

Figure 12 shows evaporator bends temperature profile with ejectors being disengaged (left) and
Figure 12
engaged shows
(right). evaporator
When bends
ejectors are not temperature profile
in operation, the bendswith ejectors being
temperatures are maldistributed
between 14°F(left)
disengaged and 25°F
and(16°F as the(right).
engaged mean mass
When weighted
ejectorsvalue), thisin
are not being explained
operation, bybends
the a
certain degree of superheating inside the coils. Superheating is typically necessary to avoid
temperatures are maldistributed between 14°F and 25°F (16°F as the mean mass
liquid return to compressors but adversely affects coil heat transfer due to poor sensible heat
weightedWhen
rejection. value), this being
ejectors explained
are engaged, bendsbytemperatures
a certain degree of superheating
are evenly inside
distributed and the
collapse
coils.
to Superheating
an average of 26°F: is
notypically necessary
sensible heat to avoid
is in place liquid return
and refrigerant latentto compressors
heat can be fully but
exploited within the coils, leading to a consistent increase in average evaporation
adversely affects coil heat transfer due to poor sensible heat rejection. When temperature
ejectors
and a strong reduction in the coil delta T. In this way, system efficiency is strongly improved, and
are engaged, bends temperatures are evenly distributed and collapse to an average
the 11% COP improvement mentioned in the previous section is therefore confirmed.
of 26°F: no sensible heat is in place and refrigerant latent heat can be fully exploited
within the coils, leading to a consistent increase in average evaporation temperature
and a strong reduction in the coil delta T. In this way, system efficiency is strongly
improved, and the 11% COP improvement mentioned in the previous section is
therefore confirmed.

Permanent Magnet Motors

To evaluate compressors’ performance improvements arising from the use of

permanent magnet motors, a CO2 transcritical test rig has been used. Its basic piping
and instrumentation (P&I diagram) is shown in Figure 13.

Technical Paper #2 © IIAR 2024 17

 Permanent Magnet Motors 2024 Natural Refrigeration Conference & Heavy Equipment Expo
To evaluate compressors’ performance improvements arising from the use of permanent
magnet motors, a CO2 transcritical test rig has been used. Its basic piping and instrumentation
(P&I diagram) is shown in Figure 13.

Figure13.
Figure [Link]
Lab test
test rigrig
P&IP&I
for for compressors
compressors performance
performance measurements.
measurements.

The following measurements were performed:

The following measurements were performed:
• p_S: suction pressure
- p_S: suction pressure
• -T_S: suction
T_S: suctiontemperature
temperature
• -m: m:refrigerant
refrigerantmass
mass flow
flow
• -P_el: power
P_el: powerconsumption
consumption
• -p_D: p_D: dischargepressure
discharge pressure
- T_S: discharge temperature
• T_S: discharge temperature
Two different compressors executions were tested: compressor A was a 50 hp CO2 transcritical
model, while compressor
Two different B was
compressors identical to
executions compressor
were A with the exception
tested: compressor of the
A was a 50 hp sole
CO2 rotor,
which was equipped
transcritical model,with
whilepermanent magnets.
compressor B wasThis kind oftotest
identical is representative
compressor A withofthe
the real
permanent
exception of magnet’s
the solebenefits, as no difference
rotor, which is in place
was equipped with between
permanent compressor
magnets.A This
and kind
compressor B from a mechanical
of test is representative standpoint
of the real (bothmagnet’s
permanent have the same components
benefits, and the same
as no difference is in
tolerances), the only difference is the rotor execution, one with and one without permanent
place between compressor A and compressor B from a mechanical standpoint (both
magnets.

18 © IIAR 2024 Technical Paper #2

CO2 Systems Add-Ons: Calculations and Field Measurements

have the same components and the same tolerances), the only difference is the rotor
execution, one with and one without permanent magnets.

A comprehensive test campaign was performed with the compressors’ performance

A comprehensive test campaign was performed with the compressors’ performance being
being compared at different operating conditions, as shown in Figure 14.
compared at different operating conditions, as shown in Figure 14.

Figure 14. Lab tests for standard motor and permanent magnet motors executions.
Figure 14. Lab tests for standard motor and permanent magnet motors executions.

Figure 14 shows that compressor B execution with permanent magnet motor significantly
Figure 14 shows that compressor B execution with permanent magnet motor
outperforms the standard execution and confirms the 9% performance improvement indicated
significantly
in the previousoutperforms
section. the standard execution and confirms the 9% performance
improvement indicated in the previous section.

Conclusions
Conclusions
Carbon dioxide is an environmentally friendly alternative to synthetic refrigerants and offers
very good performance levels in many applications. However, the typical flash-gas bypass
execution suffers from an efficiency penalty with increasing heat-sink temperatures, limiting its
Carbon dioxide is an environmentally friendly alternative to synthetic refrigerants and
widespread use in all climates. However, during the last decade, several add-ons to such
offers very
systems havegood
been performance levels
shown to perform well,infrom
many applications.
both However,
from a calculation the typical flash-
and experimental
gas bypass execution suffers from an efficiency penalty with increasing heat-sink
standpoint:
temperatures, limiting its widespread use in all climates. However, during the last
- Parallel compression
decade, several add-ons to such systems have been shown to perform well, from both
- Liquid and vapor ejectors
from- aAdiabatic
calculation and experimental standpoint:
cooling
• -Parallel compression
Permanent magnet motors
• Liquid and vapor ejectors
Depending on the given system boundary conditions (e.g., jobsite location, local humidity,
• Adiabatic
refrigeration cooling
load profile, etc.) it may make sense to implement all of them, which may bring the
• Permanent magnet
system to a COP improvement motorsabove 40% when considering the flash gas bypass execution as
the baseline.

Technical Paper #2 © IIAR 2024 19

 2024 Natural Refrigeration Conference & Heavy Equipment Expo

Depending on the given system boundary conditions (e.g., jobsite location, local
humidity, refrigeration load profile, etc.) it may make sense to implement all of them,
which may bring the system to a COP improvement above 40% when considering the
flash gas bypass execution as the baseline.

References

[1] M. S. Bhatti. 1997. “A critical look at R-744 and R-134a Mobile Air
Conditioning Systems.” Journal of Passengers Cars, Part 1, pp 181-205.

[2] K. Visser. 2018. “Moving the CO2 ‘equator’ from the northern Mediterranean to
Malaysia without ejectors.” Proceedings of IIR 2018 – International Institute of
Refrigeration – paper 0018.

[3] K. Madsen. 2018. “Financial aspects of ejector solutions in supermarkets and

smaller industrial systems.” Proceedings of IIR 2018 – International Institute of
Refrigeration – paper 0010.

[4] DORIN selection software, online version, Release 23.11, “[Link].”

[5] L. Rossi, 2018: Atmosphere China, “Architecture Sustainability – From

Switzerland to Beijing, CO2 refrigeration in a local supermarket.”

[6] Frigoconsulting, 2013, Migros supermearket in Bulle, First ejector operating

successfully, “[Link]

20 © IIAR 2024 Technical Paper #2

CO2 Systems Add-Ons: Calculations and Field Measurements

Notes:

Technical Paper #2 © IIAR 2024 21

 2024 Natural Refrigeration Conference & Heavy Equipment Expo

Notes:

22 © IIAR 2024 Technical Paper #2