4ppliedThermalEngineering Vol. 16, No. 2, pp.

103-113, 1996
ElsevierScienceLtd
Pergamon 1359-4311(95)00051-8 Printed in Great Britain
1359-4311/96 $29.00+ .00

REVIEW PAPER

VARIABLE-SPEED CAPACITY CONTROL IN REFRIGERATION

SYSTEMS

T. Q. Qureshi and S. A. T a s s o u
Department of Mechanical Engineering, Brunel University, Uxbridge, Middlesex UB8 3PH, U.K

(Receired 17 February 1995)

Abstract--This paper presents a review of the application of variable-speed capacity control to

refrigeration systems. The aim is to put together diversified information in a single source and to appraise
recent advances in variable-speed technology. The review reveals that although variable-speed drives based
on inverters have been applied successfully to control the capacity of rotodynamic machines, such as
pumps and fans, their application to positive displacement machines, such as compressors, has so far been
restricted to small-capacity air-conditioning units. There has been only a very small uptake of the
technology in the medium-range capacity units, due to a number of problems, such as insufficient
development and integration of components, poor reliability, high capital cost and the failure of
demonstration installations to produce the expected energy savings. Although inverter-based variable-
speed compressor technology offers the potential for energy savings, considerable research work is still
required for the development of optimised and cost-effective systems.

Keywords Reffigeralion. variable-speed control, m~erters

NOM ENCI. ATURE

EER energy efficiency ratio

SEER seasonal energy efficiency ratio
TEV thermostatic expansion ~alve
PWM pulse width modulation
VSI voltage source inverter
CSI current source inverter
COP coefficient of performance

INTRODUCTION

The inefficient use o f electricity to drive the c o m p r e s s o r s o f refrigeration a n d a i r - c o n d i t i o n i n g

systems is r e g a r d e d as an indirect c o n t r i b u t o r to the emission o f greenhouse gases to the
a t m o s p h e r e . These emissions can be reduced by i m p r o v i n g the energy conversion efficiency o f
refrigeration systems. One m e t h o d o f achieving this is t h r o u g h c a p a c i t y control, which m a t c h e s the
system c a p a c i t y to the load. C a p a c i t y c o n t r o l reduces the o n / o f f cycling losses o f the e q u i p m e n t
a n d improves the steady-state efficiency o f the plant due to a lower pressure differential across the
c o m p r e s s o r at p a r t - l o a d c o n d i t i o n s [1].
C o m p r e s s o r c a p a c i t y c o n t r o l techniques can be e m p l o y e d either within or outside the c o m p r e s s o r
but their basic function o f varying the refrigerant flow-rate in the cycle remains the same. C a p a c i t y
c o n t r o l m e t h o d s c o m m o n l y e m p l o y e d are: o n / o f f control, hot gas bypass, e v a p o r a t o r t e m p e r a t u r e
control, clearance v o l u m e control, multiple c o m p r e s s o r control, cylinder u n l o a d i n g a n d variable
speed control. Theoretical c o m p a r i s o n o f various capacity c o n t r o l m e t h o d s at full- and p a r t - l o a d
c o n d i t i o n s in Fig. I shows ~ariable speed as being the m o s t energy efficient technique [2, 3].
The a p p l i c a t i o n o f variable-speed capacity control to refrigeration a n d a i r - c o n d i t i o n i n g systems
has been under c o n s i d e r a t i o n over the last 20 yr. A l t h o u g h its c a p a c i t y m o d u l a t i o n features have
been p r o v e n in s m a l l - c a p a c i t y systems, there are still a n u m b e r o f p r o b l e m s which i m p e d e its
i m p l e m e n t a t i o n in medium-size systems. These p r o b l e m s include:

1. Insufficient d e v e l o p m e n t a n d integration of c o m p r e s s o r s a n d v a r i a b l e - s p e e d drives.

2. Relatively high cost due to the use of g e n e r a l - p u r p o s e v a r i a b l e - s p e e d drives.
104 T . Q . Qureshi and S. A. Tassou

Discharge gas
bypass

Suction pressure
control

Clearance volume
control

On/off control

Multiple
compressor

Variable
speed

Ideal
I I I I I I I I I I
0 10 20 30 40 50 60 70 80 90 100
Full load power (%)

Fig. I Comparison of various capacity control techniques at half load.

3. Insufficient information from manufacturers on the performance characteristics of variable-

speed systems.
4. Failure of demonstration installations to produce the expected energy savings.
5. Poor reliability of initial installations caused by unsophisticated and inadequately developed
control systems.
Research at Brunel University aims to provide answers to some of the above uncertainties, to
identify the most suitable compressor and VSD technologies for variable-speed operation and to
develop control strategies for the cost-effective integration of these technologies.
This paper presents an in-depth look into the whole subject by compiling previous published
information on the development of the technology. The review addresses both the mechanical and
electrical aspects of variable-speed capacity control.

R E F R I G E R A T I O N CAPACITY C O N T R O L T H R O U G H C O M P R E S S O R SPEED
MODULATION

The basic difference between variable-speed refrigeration and conventional refrigeration systems
is in the control of the system capacity at part-load conditions. In variable-speed refrigeration the
capacity of the refrigeration system is matched to the load by regulating the speed of the compressor
motor in such a way that the capacity of the system tracks the load dictated on it by varying
operating conditions. Variable-speed control can be realised in a number of ways, which can be
divided into two groups. Firstly, those in which the load is indirectly coupled to the motor (a
constant-speed motor and a speed-control device between the motor and the load) and, secondly,
those in which the load is directly coupled to the motor (a variable-speed motor). The first group
further subdivides into mechanical, hydraulic and electrical systems and the second group into d.c.
drives, switched reluctance drives, multi-speed motors, Ward Leonard sets and electrical variable-
speed drives. The advantages and disadvantages of each group of capacity control methods have
been discussed in greater detail in ref. [4]. In refrigeration applications, stepwise or infinitely
variable control of the motor speed are considered to be the most flexible methods of speed control
and significant energy savings have been reported through their application [5, 7-10].
Stepwise speed control can be achieved by using multi-pole electric motors. The required
compressor capacity is obtained by switching a finite number of poles to achieve the desired speed.
 Variable-speed control in refrigeration systems 105

Stepwise control is less costly than continuous speed control but step-controlled motors have lower
- efficiency than constant-speed motors [3]. Moreover, this method of speed control has the limitation
of a fixed number of speeds, which offers restricted compressor capacity control compared to
continuous stepless capacity control.
Infinitely variable capacity control can be realised by using electronic variable-speed drives to
regulate the speed of the compressor motor. Since the torque-speed performance characteristics
of an induction motor at low-speed operation are the same as those at rated motor frequency,
frequency variation is considered to be an efficient speed-control technique.

R E S E A R C H INTO T H E M E C H A N I C A L ASPECTS OF V A R I A B L E - S P E E D
REFRIGERATION

Early work on variable-speed refrigeration systems was directed towards the theoretical analysis
of the concept of variable-speed capacity control and the investigation of the problems associated
with the mechanical design of the system. Most of' the published work discusses the overall
performance and benefits of the system, rather than the establishment of criteria for the integration
and optimisation of compressors and variable-speed drives.
Cawley et al. [5] compared the part-load efficiency of two-speed compressors with compressor
unloading capacity control. It was found that 49% better energy efficiency ratio (EER) could be
realised by a system with a two-speed compressor, compared with a system using a cylinder
unloaded compressor. The basic reason stated for this improvement was decreased power input
requirements in the two-speed compressor, due to lower frictional losses at half speed.
Cohen et al. [6] analysed the energy conservation potential of variable-capacity compressors in
domestic and small, commercial, air-conditioning systems. The authors reported the effects of using
variable capacity control and identified essential modifications for the attainment of maximum
efficiency gains. Although no specific technique For displacement control was recommended, it was
emphasised that energy could be saved on a seasonal basis because the system would operate more
efficiently at lower capacities, due to the reduced frictional losses in the compressor. Energy savings
were also anticipated due to the reduced pressure ratio imposed on the compressor by the lower
temperature difference at lower loads in both condenser and evaporator. Efficient refrigerant
control devices which could quickly respond to wide variations of pressure ratio were also
recommended for variable-capacity systems because the response of capillary tubes was found to
be unsatisfactory compared to that of thermostatic expansion valves (TEVs). It was concluded that
variable capacity control could provide energy savings of 28 35% on a seasonal basis, without
significant changes in the system size and cost. Other projected advantages were reduced torque
requirements at initial start up, low noise operation at most loads, fewer on/off cycles and stable
humidity control. The investigations did not consider the practical implications of low capacities
and low flow-rates on the design of the refrigerant piping and the heat exchangers.
Muir e t al. [7] investigated various aspects of capacity modulation for general refrigeration and
domestic air-conditioning systems using a rating technique, the seasonal energy efficiency ratio
(SEER). This method compares the seasonal efficiency of domestic air-conditioning systems, taking
into consideration the effects of on/off cycling and steady-state efficiency at several outdoor
temperatures. The analysis showed that significant energy savings would be possible through
capacity modulation, due to a decrease in on/off cycling losses and improvements in steady-state
efficiency at lower loads than the design load. It was concluded that, in addition to investigating
the costs and savings that could arise from speed modulation, more research was needed to
determine the on/off cycling rates, load/capacity ratio and on/off cycling degradation coefficients
for different applications.
Lida et al. [8] carried-out experimental investigations on a heat pump equipped with a 4 hp,
(3 kW), hermetic rotary compressor. It was found that the practical limits for compressor speed
variation were between 25 and 75 Hz. The results indicated improvements in EER with the
inverter-driven compressor, compared to a fixed-capacity system. The reason stated for the
improvement was higher efficiency at part load, which reduced the power consumption and cycling
losses. Cost and SEER analyses showed a 20% increase in the total cost for the inverter-controlled
system and between 20 and 26% energy savings over the single-capacity system. The cycling losses
106 T.Q. Qureshi and S. A. Tassou

were estimated to be between 5 and 7% and the payback period was calculated to be between 3
and 4 yr. Other advantages identified for variable-speed control over fixed-speed systems included
accurate temperature control, system soft-start capabilities and low noise operation at reduced
loads.
The work of Shimma et al. [9] concentrated on the evaluation of energy savings from the
application of inverters to air-conditioners and considered the effects of employing inverters in some
detail. The authors proposed that maximum energy savings and better system performance could
be achieved by employing better control methods and improving the performance of individual
components in the air-conditioning system. Anticipated energy savings for small-capacity inverter-
driven air-conditioners were in the range between 20 and 40%. These savings were attributed
mainly to higher operating efficiency at lower speeds. The refrigeration effect in the low-frequency
region was found to increase because of the low compression ratio operation which results from
increased heat-exchanger capacity. A microcomputer-based control system was used to control the
room temperature. The optimum operating frequency of the compressor was determined by the
microcomputer with input values from the room-temperature sensor and heat-exchanger tempera-
ture sensor using P| (proportional and integral) control logic. The capacity-controlled system
resulted in a reduction of the room temperature fluctuations to 50% of those for the conventional
on/off-controlled system. In addition to this, it has been envisaged that the time required to bring
the room temperature to the desired set-point temperature could be reduced by running the
compressor at a higher speed at start-up. The authors pointed out various problems that needed
further consideration. These included improvements in the refrigerant throttling mechanism,
adoption of more effective noise suppression techniques to reduce radio wave interference noise
and harmonic noise generated by the inverter, enhancement of the reliability and performance of
the inverter, improvements in the overall system design to reduce noise at high-frequency operation
and to overcome vibration problems at low-frequency operation. It was concluded that the cost
of the inverter control system needed to be reduced further to expand its application.
Itami et al. [10] examined ,qe performance and reliability factor of frequency-controlled,
reciprocating and rotary compressors of the rolling piston type. They suggested modifications to
ensure reliability which differ for each type of compressor. For instance, with the reciprocating
compressor, a two-stage oil-pump was used over the low-frequency range to ensure proper
lubrication. For the rotary compressor, a liquid injection system was used to limit overheating and
a disc mechanism was adopted to prevent increased amounts of discharge oil at the higher
operating-frequency range. The rotary compressor showed improvements in the volumetric and
motor efficiencies when the operating frequency was increased, whereas the reciprocating compres-
sor exhibited improvement in mechanical and compression efficiencies when the operating
frequency was decreased. Between 20 and 40% improvement in the SEER was reported with the
frequency-controlled air-conditioner, compared to the conventional on/off-controlled system.
In recent years, scroll compressor technology has shown promising efficiency advantages over
comparable positive displacement compressors. This is due to the smooth and continuous
compression characteristics of the scroll design and the elimination of valve losses. Ischii et al.
[19, 20] compared the mechanical efficiency and dynamic performance characteristics of scroll
compressors with those of roiling-piston rotary compressors. It was found that the scroll
compressors exhibited better vibration characteristics than the rolling-piston rotary compressor but
lower mechanical efficiency. It was anticipated that the mechanical efficiency of scroll compressors
could be improved through design optimisation.
The investigations of Senshu et al. [21] on a small-capacity heat-pump employing a scroll
compressor showed a 30% improvement in annual performance efficiency, compared with the
conventional reciprocating compressor. The EER of the inverter-driven heat-pump at nominal load
conditions, however, was found to be less than that of a constant-speed system, due to the inverter
losses.
A feasibility and design study of a continuously variable capacity refrigeration system was carried
out under the Energy Efficiency Demonstration Scheme on behalf of the Department of Energy
[22]. A commercially available variable-speed system was monitored in a supermarket application
with a view to first assessing the performance of an already installed conventional system and then
converting these units to variable speed for overall comparison. The investigation showed a 56%
 Variable-speed control in refrigerationsystems 107

power saving with high temperature (dairy applications) and a 30% saving with low temperature
(frozen-food applications). The energy savings achieved were attributed mainly to variable-speed
control and fully floating head pressure.
Rice et al. [23-25] reported energy savings of the order of 27% for a modulating heat-pump
system arising from reduced cycling losses, heat-exchanger unloading, reduced frosting/defrosting
losses and reduced back-up heating. It was found that increased motor-slip losses and distorted
inverter waveform decreased the conventional three-phase induction motor efficiency by up to 20%,
depending on frequency and inverter type. It was suggested that a permanent-magnet, electronically
commutated motor-inverter combination could reduce these losses.
ASHRAE research project RP-409 analysed a large chiller employing a variable-speed-controlled
centrifugal compressor [11]. The results showed that variable-speed control led to a 1.5% reduction
in the compressor power consumption at maximum load and a 40% reduction at minimum load.
Wong et al. [12] confirmed by experimental investigation that variable compressor speed control
is more efficient compared to cylinder unloading control. With variable speed, volumetric and
isentropic efficiencies and COP increased when the compressor speed was reduced, while cylinder
unloading control exhibited reduced isentropic efficiency and COP. The authors evaluated the
economic benefits of a variable-speed compressor in another research paper [13]. It was shown that
variable-speed control leads to reduced energy consumption, but for intermittent operation it may
not be economically viable due to the high capital cost of the inverter.
The work of Tassou et al. [1, I~17] concentrated mainly on the capacity control of domestic-size
heat-pump systems. Important issues investigated included energy conservation with capacity
control, performance comparison with conventional systems, effects of capacity modulation,
mathematical modelling of variable-speed systems, part-load and dynamic performance analysis of
heat pumps. The investigations showed that variable-speed control could achieve a 15% improve-
ment in energy conversion efficiency, compared to a conventional system. It was also found that
superheat control with a thermostatic expansion valve was unsatisfactory during part-load
operation and it was suggested that the problem could be effectively overcome by employing a
microprocessor-controlled motorised expansion valve.
McGovern [18] investigated the performance of a two-cylinder, open-type, reciprocating
compressor over the speed range 300-900 rpm. Performance parameters, such as mass flow-rate,
shaft power and compressor discharge gas temperature showed a linear increase over the tested
speed-range, whereas the volumetric efficiency was found to remain almost constant at about 66%
over the speed range. The variation in mechanical efficiency with speed was found to be very small,
increasing from 92 to 94% as the speed increased from 300 to 900 rpm.

E L E C T R I C A L ASPECTS OF V A R I A B L E - S P E E D R E F R I G E R A T I O N
Inverter-based variable-speed drive technology is presently well proven in various applications
throughout all sectors of industry and several drive types are available for both energy conservation
and high-performance applications. A VSD is an interface between the utility input and the
compressor motor that controls the speed of the motor by changing the magnitude of voltage,
current or frequency. A three-phase VSD or frequency converter, shown in Fig. 2, consists of a
rectifier which converts the three-phase mains voltage, i.e. 415 V, 50 Hz, to a d.c. voltage either
controlled or uncontrolled, and an inverter which inverts the d.c. voltage to a three-phase a.c.
supply-voltage to the compressor motor. The voltage at the inverter output is adjustable in
magnitude and frequency. A regulator is used to change the inverter switching characteristics so
that the output frequency can be controlled. This may comprise a sensor which measures the
control variable and sends a feedback signal to the system.

Classification o f electronic VSDs

A basic classification diagram of VSDs is shown in Fig. 3. The classification criteria used are
the type of semiconductor switch employed, input and output circuit topology, motor type, control
strategy, power and voltage level, and regeneration capability [26]. A comprehensive technology
review of the application of power electronics is given by Bose [27], which includes a discussion
of power semiconductor devices, power integrated circuits, converter circuits and applications.
108 T.Q. Qureshi and S. A. Tassou

Variable voltage and

Input power variable frequency output
supply _l
Frequency] / ~
21 converter I
21 and inverter I
3-Phase 50Hz -I I
a.c. motor Compressor

d.c, link 1
3-Phase input _1 1 ÷_ I I : 3-Pha~ output
:,21 .e .,or I _-I ,nverter ,.I ; --

Constant voltage -I I --I I : v~ablo voltagean~

and constant Converter section Inverter section
| .
,, variable frequency
frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i

Variable speed drive

Fig. 2. Basic configuration of electronic variable-speed drive.

Currently available VSD systems can be classified into three basic inverter types: the six-step
voltage inverter (VSI), the six-step current inverter (CSI) and the pulse-width-modulated voltage-
source inverter (PWM). All three types have certain relative advantages and disadvantages and it
is unlikely that any one will displace the others in all types of application [28]. A recent survey,
however, has shown that over the last few years sales of P W M inverters exceeded those of other
inverter types [29]. A report published by the Energy Efficiency Office c o m p a r e s typical efficiencies
of six V S D types of different ratings. As shown in Fig. 4, the P W M inverter shows a slightly better
efficiency over VSI and CSI [30].
The efficiency of the inverter and the m o t o r greatly influence the energy-efficiency ratio of the
refrigeration system. The efficiency of a typical inverter decreases with a decrease in speed. The
combined efficiency of a conventional inverter and m o t o r is around 86% at m a x i m u m speed and

Variable-s eed drive

Input Inverter
converter

I
Constant Variable Voltage Current
voltage voltage switching switching
converter converter inverter inverter

I I I I
Pulse-width Six-step Six-step Load
modulated voltage current commuted
source source current
PWM inverter inverter inverter
VSI CSI LCI

Fig. 3. Classification of electronic variable-speed drives.

 Variable-speed control in refrigeration systems 109

50% Speed-25 % torque

50% Speed-100% torque
100% Speed-100% torque

Multiple-speed
motors

Eddy-current
couplings

Switched
reluctance
drives

Pulse-width
modulated

Current-source
inverter

Voltage-source
inverter

0 !0 20 30 40 50 60 70 80 90 100
Full load efficiency (%)

Fig. 4 Typical efficiency of various VSD types [36].

full load and falls off at reduced loads and speeds [31]. Since refrigeration systems operate most
of the time at part-load conditions, considerable energy savings can be achieved by raising the
efficiency of the inverter and motor combination at part-load/low-speed operation. Alternately,
selection of these main driving units should be such that they operate in the high-efficiency region
for most of their duty cycle.

Effects of electronic VSDs

A variable-speed inverter is liable to introduce disturbances, especially harmonics of the supply
frequency, into the system to which it is connected, due to the effects of the switching mode of
operation.
The negative effects of VSDs can be classified into two categories: those on the supply side and
those on the motor side. The main effect on the supply side is radio-frequency interference, which
may cause disturbances in the communication equipment in the vicinity of the inverter. On the
motor side, the effect could be increased motor losses due to a non-sinusoidal voltage waveform,
speed/torque oscillations which impose extra stresses on the windings and variation in slip which
directly affects the torque of the motor.
Lloyd [32] investigated the effects of various waveforms on the efficiency of the motor and
compressor. The waveforms analysed were six-step and pulse-width modulation (PWM), which
were compared with pure sinusoidal waveform. The sinusoidal waveform showed higher motor and
compressor efficiencies over the entire frequency range, whereas the PWM waveform resulted in
a better motor and compressor efficiency over the six-step waveform at higher frequencies.
Mohen et al. [33] evaluated the feasibility of heat-pump capacity modulation with compressor
speed adjustment using six different adjustable speed techniques. PWM and VSI drives were found
to be more efficient than other adjustable-speed motor drive methods such as: square wave current
source inverter (CSI), electronically-commutated synchronous motors with a permanent magnetic
field (ECM), high-frequency high-speed motors using low-loss magnetic material (HCM) and pole
amplitude-modulation motors (PAM). These techniques are capable of providing speed modu-
lation in the range of at least 6: 1, allowing speed to be varied above and below the rated running
speed. The power factor of the PWM inverter was found to vary in the range between 60 and 80%,
110 T.Q. Qureshi and S. A. Tassou

which is generally considered to be within permissible limits, whereas for the VSI the power factor
decreased to an unacceptable value, down to 10% at low-speed operation. The efficiency of both
drives was estimated to be 95% at full load. It was found that both inverter drives introduced large
amounts of harmonic current into the utility system and modifications to the relevant circuitry to
eradicate or lessen the harmonic current and to enhance the power factor of operation were
suggested. The input distortion index was found to be in the range between 70 and 130% for both
inverters.
Scholey [34] compared the performance characteristics of sine wave, six-step square wave and
pulse-width-modulated variable-frequency power supplies. The induction-motor performance
when subjected to complex voltage waveforms and variable-frequency supplies was also evaluated,
considering winding stresses, low-speed pulsating torques, temperature rise and the effect of
harmonics. It was found that the voltage-source inverter produced losses of 20%, compared to the
pure sine-wave supply. The losses with the PWM inverter were not quantified but stated as being
less than the VSI and dependent on the switching frequency. It was also mentioned that
low-frequency pulsating torques have less effect on the PWM inverter than on the six-step VSI,
due to the large number of pulses per cycle, which favour PWM when stable operation at low speeds
is required. Motor temperature-rise was stated as being excessive at lower speeds for both types
of inverter, due to the reduction in ventilation and additional harmonics.
Rahman [35] reviewed the efficiency of an inverter-fed induction-motor and identified various
losses which may occur in the motor due to harmonics, which include copper losses, stator losses,
core losses and stray losses. The power loss caused by the inverter was shown to increase with the
increase in the power rating of the drive.
There are various standards that deal with the disturbances in the supply system and provide
guidelines for harmonic control. The most coJ..monly referred to standards are:
• Engineering recommendation G.5'3. "Limits of harmonics in the U K electricity supply system'
[36].
• IEEE Standard 519-1981, 'IEEE guide for harmonic control and reactive compensation of
static power converters' [37].
• International Electrotechnical Commission Standard, IEC 555, 1982 [38].
• British Standard BS 5406, 1998, 'Disturbances in supply systems caused by household
appliances and similar electrical equipment [39].
Harmonic currents can be controlled by various methods, such as shunt filters, phase multipli-
cation and harmonic compensation or injection. A correctly designed input-filter can reduce the
dominant harmonic to the acceptable level.

A.C. MOTORS

The type of motor fitted to the compressor is a primary consideration in selecting an electronic
VSD. A number of choices are available, including both synchronous and asynchronous types, but
the standard induction motor is the most widely used due to cost, reliability and availability
advantages over other motor types.
According to Domijan [29], the efficiency of an inverter-fed induction-motor compared to a direct
mains-driven motor is higher from start-up to no-load conditions, but decreases under part- and
full-load conditions, due to increased motor losses in the presence of multiple harmonics in the
current and voltage waveforms. These losses must be taken into consideration in motor design,
together with the influence of motor temperature-rise, low-frequency pulsating torques, voltage
stresses and harmonic torques.
High efficiency motors are a new development and achieve maximum efficiency at between 70
and 85% of full load [40]. The efficiencies of high-efficiency motors and standard induction-motors
under various loading conditions are compared in Fig. 5. It can be seen that high-efficiency motors
operate at maximum efficiency over a wider range of loads compared to standard induction-motors
and their peak efficiency is approximately 5% higher than that of standard induction-motors, due
to lower component losses. Motors utilising a higher class H insulation system are now specially
designed for variable-speed operation.
 Variable-speed control in refrigeration systems 111

REFRIGERATION COMPRESSOR

Currently, various compressor types are being driven by inverters and some are better suited to
VSD operation than others, due to their better performance characteristics. The main types of
positive displacement compressors likely to be used in commercial refrigeration and air-condition-
ing applications are reciprocating, rotary, screw and scroll. The selection of a compressor for a
particular application incorporating a VSD is not quite as straight-forward as it may seem. System
design has to take into account the anticipated load-profile at variable-speed operating conditions.
The subject has been discussed in more detail by Mills [41]. Some of the basic requirements of a
compressor for VSD applications are [42]:
(i) Proper lubrication at low- and high-speeds must be ensured as inadequate lubrication at
low-speeds may increase overheating and friction losses of compressor components and
ample lubrication at high-speeds may damage the seals and gaskets. An improperly designed
lubrication system may reduce the performance, reliability and life of a compressor.
(ii) The compressor support frame should be designed such that the resonance frequencies are
above the operating frequency range. A compressor running at fixed speed imposes vibration
on its framework at a set group of frequencies. The framework is normally designed such
that its natural frequency differs from the imposed frequency. A variable-speed compressor
design will be more complex as each speed will impose different frequencies.
(iii) The VSD should not increase the stresses on the suction and discharge valves. Compressor
valves designed to operate at fixed speed may not be suitable for variable-speed operation
as imperfect valve action at various speeds may increase valve inefficiencies.
(iv) The capacity of the compressor should vary in direct proportion to speed and the efficiency
of the compressor should not decrease within the required speed range.
Thermodynamically all compressor types follow the same laws but their efficiencies can vary
considerably from one type to another.
Riegger [43] compared the performance of commercially available small-capacity rotary,
reciprocating and scroll compressors to evaluate their performance under variable-speed operation.
It was found that all three compressors were optimised for 60 Hz operation and their energy-
efficiency ratio decreased above and below this rated point. The authors concluded that there is
no straight-forward answer to the question of which type of compressor is most suitable for
variable-speed operation because various factors, such as capacity range, operating conditions and
manufacturing cost, influence their seasonal energy efficiency.
Tassou et al. [44] compared the performance of open-type reciprocating, semi-hermetic recipro-
cating and rotary vane compressors over a range of speeds and loads using a 25 kW nominal

95 • - - - | - -

90 ~ ~ f..--,---- I~

85,~"
/
80
• Energy-efficient m o t o r

?5
| i
* Conventional induction-motor
I I l I ] i
25 35 45 55 65 75 85 95
Load (%)

Fig. 5. Efficiencycomparison of high-efficencymotor and conventional induction-motor.

112 T.Q. Qureshi and S. A. Tassou

capacity refrigeration system. Of the three compressors, only the open-type reciprocating unit
exhibited an appreciable increase in the COP with a reduction in speed. The semi-hermetic
reciprocating compressor exhibited only a small improvement, due to the negative effects of reduced
motor cooling and high suction temperatures, whereas the rotary compressor exhibited a reduction
in the COP at low speeds, due to reduced hydrodynamic sealing between the rotating vanes and
the cylinder. An analysis of the seasonal energy performance of a refrigeration system equipped
with an open-type compressor and used for air-conditioning applications has shown that the system
can lead to between 12 and 24% energy savings, compared to a conventional fixed-speed system,
depending on climatic conditions [45].

CONCLUSIONS

In recent years, scroll compressors with VSD drives have gained an increased market share in
small packaged air-conditioning systems, due to improved efficiency and reliability brought about
by advancements in material and manufacturing technology and optimised design. In the
medium-capacity range, reciprocating, rotary and scroll compressors are being used with off-the-
shelf inverters. Research results have shown that open-drive compressors lead to better efficiency
and allow smaller condensers to be fitted as the increased motor heat caused by the inverter losses
is not rejected in the condenser.
Over the past few years, the price of inverters has been steadily decreasing. This trend is
attributed generally to improved design, increased production and the decrease in the price of
power electronic devices. The reliability of inverters is also constantly improving, due to improved
technology and the availability of better components. Space requirements are also decreasing, due
to improved packaging, higher efficiency, more effective heat sinks and improved circuitry.
The downward price trend and new technological developments are favourable for the increased
use of VSD refrigeration in place of conventional refrigeration systems. There is also scope for
achieving further energy savings by using high-efficiency motors if their present costs become
competitive with standard induction-motors.
The development of an optimum variable-speed refrigeration system is a function of several
design factors and more research work is needed to fully understand the interaction of the
components in an integrated VSD refrigeration system. Problems to overcome are the generation
of harmonics by the inverter, which affects both the supply and the motor, and the reduction of
the motor efficiency at low speeds. Proper lubrication and cooling of the compressor at low speeds
is also an important consideration.
New developments in inverter technology, such as vector and fuzzy logic control, require further
investigation as to their application to refrigeration systems. The energy-efficiency of optimised
VSD refrigeration systems should also be investigated in different applications, such as low-tem-
perature, medium-temperature and high-temperature refrigeration.

REFERENCES

I. S. A. Tassou, C. J Marquand and D. R. Wilson, Comparison of the performance of capacity-controlled and

conventional-controlled heat-pumps. Appl. Energy 14, 241 256 (1988).
2. H. Janssen and H. Kruse, Continuous and discontinuous capacity control for high-speed refrigeration compressors.
Proe. Purdue Comp. Tech. Con£, Purdue University, USA, 1984.
3. S. M. Zubair and V. Bahel, Compressor capacity modulation schemes. Heating, Piping, Air-cond.. January, 135-143
(1989).
4. T. Sheldrake, Introducing variable speed drives. Building Services, April, 25-32 (1991).
5. R. E. Cawley and D. M. Pfarrer, Part-load Efficiency Advantages of Two-speed Refrigerant Compressors, pp. 42~46.
6. R. Cohen, J. F. Hamilton and J. T. Pearson, Possible energy conservation through the use of variable-capacity
compressor. Proe. Purdue Compressor Technology ConJl, Purdue, USA, 1974, pp. 5(~54.
7. E. B. Muir and R. W. Griffith, Capacity modulation for air-conditioner and refrigeration system. Air-cond., Heating
Refrig. News, April, 3 16 (1979).
8. K. Lida, T. Yamamoto, T. Kuroda and H. Hibi, Development of an energy-saving-oriented variable-capacity system
heat-pump. ASHRAE Trans. 88, 441~49 (1982).
9. Y. Shimma, T. Tateuchi and H. Sugiura, Inverter control systems in a residential heat-pump air-conditioner. ASHRAE
Trans., Paper HI-85-31, No. 2, pp. 1541-1552 (1988).
10. T. Itami, K. Okoma and K. Misawa, An experimental study of frequency-controlled compressors. Proc. Purdue Comp.
Tech. Conf., Purdue, USA, 1982, pp. 297-304.
 Variable-speed control in refrigeration systems 113

11. ASHRAE Research Note 63, Analysis of energy use and control characteristics of a large variable-speed drive chiller
system. ASHRAE J., January, 33-34 (1985).
12. A. K. Wong and R. W. James, Capacity control of a refrigeration system using a variable-speed compressor, Building
Serv. Engng Res. Technol. 9(2), 63~8 (1988).
13. A. K. Wong and R. C. Legg, Variable compressor speed control: economic evaluation. Building Serv. Engng Res.
Technol. 10(1), 21 27 (1989).
14. S. A. Tassou, R. K. Green and D. R. Wilson, Energy conservation through the use of capacity control in heat pumps.
J. Inst. Energy 54, 30-34 (1981 ).
15. S. A. Tassou, C. J. Marquand and D. R. Wilson, The effect of capacity modulation on the performance of vapour
compression heat pump system. International Syrup. on the Industrial Application of Heat Pumps, UK, 1982,
pp. 187 195.
16. S. A. Tassou, C. J. Marquand and D. R. Wilson, Comparison of the performance of capacity-controlled and
conventional on/off-controlled heat-pumps. Appl. Energy 14, 241 256 (1983).
17. S. A. Tassou, Experimental investigation of the dynamic performance of variable-speed heat-pumps. J. Inst. Energy
64, 95 98 (1991).
18. J. A. McGovern, Performance characteristics of a reciprocating refrigerant compressor over a range of speeds. Proc.
Purdue Comp. Tech. Conj., Purdue, USA, 1988, Vol. [, pp. 146 153.
[9. N. Ischii, M. Yamamura, H. Morokoski. M. Fukushima, S. Yamamoto and M. Sakai, On the superior dynamic
behaviour of a variable rotating speed scroll compressor. Proc. Purdue Comp. Tech, Conf., Purdue, USA, 1988,
pp. 75 82.
20. N. Ischii. M. Yamamura, S. Muramatsy, S. Yamamoto and M. Sakai, Mechanical efficiency of a variable speed scroll
compressor. Proc. Comp. Tech. Con£. Purdue, USA, 1990, Vol. I, pp. 192- 199.
21. T. Senshu, A. Arai, K. Oguni and F. Harada, Annual energy saving effect of capacity-modulated air-conditioner
equipped with inverter-driven scroll compressor. ASHRAE Trans. 91, 1569 1584 (1985).
22. UK Dept of Energy, Energy Efficiency Best Practice Programme, General Report 3, Feasibility and Design Stud), o f
Continuously- Variable Capacity Refrigeration Phmt. Dept of Energy, UK ( 1991 ).
23. C. K. Rice and S. K. Fischer, A comparative analysis of single and continuously variable-capacity heat-pump concepts.
Proc. of the DOE/ORNL Heat Pump Con]) Research mzd Detelopment on the Heat Pumps .]'or Space Conditioning
Applications, CONF-841231, Washington, 1984, pp. 57 65.
24. C. K. Rice, Efficiency characteristics of speed-modulated drives at predicted torque conditions for air-to-air-heat-
pumps. ASHRAE Trans. 94, 892 921 (1988).
25. C. K. Rice, Benchmark performance analysis of an ECM-modulated air-to-air heat-pump with a reciprocating
compressor. A S H R A E Trans. 98, 430 450.
26. D. Jarc and D. W. Novotny, A graphical approach to a.c. drive classification. IEEE Trans. Ind. Appl. IA-23, 1029-1035
(1987).
27. K. B. Bose, Power electronics--a technology review. Proc. IEEE 80, 13034334 (1992).
28. D. W. Novotny, A comparative study of variable frequency drives for energy-conservation applications, Dept of
Electrical and Computer Engineering, University of Wisconsin-Madison, USA.
29. A. Domijan and E. Embriz-Santander, Measurement of electrical power inputs to variable-speed motors and their
solid-state power converters. ASHRAE Trans. 99, 241 258 11993).
30. UK Dept of Energy, Retrofitting a.c. Variable-Speed l)rire~, Good Practice Guide 14. Energy Efficiency Office, Dept.
of Energy UK, 1991.
31. Rotocold Compressors, Technical Guide. Rotocold Ltd, Herelbrd. U K (1987).
32. J. D. Lloyd, Variable-speed compressor motors operated on inverters. ASHRAE Trans. 88, 633641 (1982).
33. N. Mohan and J. W. Ramsey, Comparative study of adjustable speed drives for heat pumps, Report No. EPRI
EM-4704, Electric Power Research Institute (1986).
34. D. Scholey, Induction motors for variable frequency power supplies. IEEE Trans. Ind. Appl. 1A-18, 368 372 (1982).
35. M. A. Rahman, Efficiency of lm,erter-]ed Induction-motors, pp. 1286~1290. IEE (1980).
36. The Electricity Council, Engineering recommendation G53. Limits for harmonics in the United Kingdom electricity
supply system, September 1976.
37. IEEE Standard 519-1981, IEEE Guide for harmonic control and reactive compensation of static power converters
(1981).
38. International Electrotechnical Commission, Standard IEC 555 I1982).
39. British Standard Institution, Disturbances in supply systems caused by household appliances and similar electrical
equipment, BS 5406, Parts 1, 2 and 3 {1988).
40. S. R. Colby and D. L Flora, Measured efficiency of high efficiency and standard induction motors, IEEE Trans., May,
18 23 (1990).
41. M. Mills, Variable-speed drives, Parts I Part 4. Re/rig.. Air-cond., Heat Recovery, February-May (1987).
42. Energy Efficiency Office, Guidance Notes Jbr Reducing Energy Consumption Costs of Electric Motors and Drive Systems,
Good Practice Guide 2. Dept of Energy, UK (1989).
43. O. K. Riegger, Variable-speed compressor performance. ASHRAE Trans. 94, 1215 1228 (1988).
44. S. A. Tassou and T. Q. Qureshi, Investigation into alternative compressor technologies for variable speed refrigeration
applications. Proc. 12th International Compressor Technology Conj., Purdue University, USA, 19--22 July 1994.
45. T. Q. Qureshi and S. A. Tassou. Energy performance evaluation of variable speed compressor refrigeration systems
in air-conditioning applications. Proc. CIBSE National CmTIi, Brighton, UK, 1994, Vol. 1, pp. 173-178.