Iran J Sci Technol Trans Mech Eng (2016) 40:325–335

DOI 10.1007/s40997-016-0028-2

RESEARCH PAPER

Optimal Charge Amount for Different Refrigerants

in Air-to-Water Heat Pumps
Faraz Afshari1 • Omer Comakli1 • Nesrin Adiguzel2 • Sendogan Karagoz1

Received: 27 January 2016 / Accepted: 10 July 2016 / Published online: 19 July 2016
 Shiraz University 2016

Abstract This paper describes an experimental study on W_ Compressor power (W)

the heat pump performance and compressor suction pres- T Temperature (C)
sure, in order to find out the optimal refrigerant charge Cp Specific heat capacity (J kg-1 K-1)
amount for different refrigerants. Independent of the choice h Refrigerant enthalpy (kJ kg-1)
of refrigerant, environmental and safety problems can be Q_ Heat transfer rate (W)
reduced by detecting the optimal amount of refrigerant AWHP Air-to-water heat pump
charge in the heat pumps or refrigeration systems. In this I Electric current (A)
investigation, a laboratory air-to-water heat pump was built U Voltage (V)
to find out the charge effects on the coefficient of perfor- m_ Mass flow rate (kg s-1)
mance and heating capacity for three different gases k Thermal conductivity (W m-1 k-1)
including R22, R134a and R404a. Gas injection was done g Gravity (m s-2)
gradually into the system, and results were obtained and hc Heat transfer coefficient (w m-1k-1)
recorded at each stage. It was shown that refrigerant D Condenser diameter (m)
overcharge and undercharge decrease performance of the r Condenser radius (m)
unit. Although heating capacity increases gradually, com- L Condenser length (m)
pressor work increases with additional charge and there B Coefficient of thermal expansion (k-1)
must be an optimum charge amount to obtain the maximum v Kinematic viscosity (m2 s-1)
coefficient of performance. e_ Energy loss (W)

Keywords Heat pump  Refrigerant  Charge Subscripts

optimization  COP a, e Air passing through evaporator
w, c Water passing through condenser
List of symbols evap Evaporator
COP Coefficient of performance comp Compressor
GWP Global warming potential cond Condenser
ODP Ozone depletion potential r Refrigerant
p Pressure (kPa) a Air
w Water
HP Heat pump
& Faraz Afshari
[email protected]
as Asbestos
o Outer
1
Department of Mechanical Engineering, Atatürk University, i Inner
25240 Erzurum, Turkey s Condenser surface
2
Department of Mechanical Engineering, Kafkas University, in Inlet
Kars, Turkey

123
326 Iran J Sci Technol Trans Mech Eng (2016) 40:325–335

out Outlet Exergy efficiency of heat pump units is significant

T Total parameter and a part of thermodynamics calculations
which has been analyzed in a heat pump cycle (Yan et al.
2016). A number of studies aimed at reducing the gas
amount in refrigerators and heat pumps. On a global level,
1 Introduction
reduction in refrigerant charge must not affect the energy
aspects while respecting environmental constraints. In the
In recent years, renewable heat technologies, including
devices working with refrigerant gas, independently of the
solar heating, renewable biofuels, geothermal heating, heat
choice of refrigerant, safety and environmental problems
pumps and heat exchangers recovering lost heat, have
can be minimized by reducing the charge amount (Fer-
received remarkable attention for both domestic and
nando et al. 2004; Poggi et al. 2008; Choi et al. 2012). It
industrial applications (Comaklı et al. 1996). Faster
has been found that the refrigerant charge in the system has
urbanization after industrial revolution brought the idea to
a considerable effect on the thermal performance and
find new ways for energy needs. The first-time district
heating capacity in the other types like direct expansion
heating system was built in New York, USA, in 1877.
solar-assisted heat pumps (Zhang et al. 2014). In a reported
Thereafter, it has been spread to most of the countries in
study, COP and heating capacity were measured for a CO2
Europe. At the beginning of the twentieth century, these
heat pump and analyzed by varying the refrigerant charge
kinds of heating models have been used in Denmark,
amount at standard cooling condition. The results presented
Germany, Belgium, Netherland, and particularly in Russia
a great variation on the performance according to the
and Sweden (Comaklı et al. 2004). In a recent study, results
refrigerant charge (Cho et al. 2005). The refrigerant charge
show that the water-to-air-type heat pump unit has the
amount is a key factor for heat pump system optimization
maximum coefficient of performance (COP) followed by
and normally determines the condensing and evaporating
water to water, air to air and air to water type. In Turkey,
pressures, which affects the subcooling, subheating and
especially in cold cities such as Erzurum, there are several
also heating and cooling capacity of the condenser and
studies to investigate heat pump’s performance due to
evaporator. Heating capacity increases as subcooling
climatic conditions and various seasons of the year. In a
increases; however, there is an optimum charge amount for
study, an experimental setup was constructed to analyze the
the best coefficient of performance (COP) (Corberan et al.
performance of the solar-ground source heat pump. The
2008). Optimal refrigerant charge of a water-to-water
experimentally obtained results were used to find the heat
propane heat pump (Corberan et al. 2011), the effect of
pump coefficient of performance (COP) and the total per-
refrigerant charge amount on single and cascade cycle heat
formance of the system (COPT), which were calculated in
pump systems (Kim et al. 2014), refrigeration systems with
the range of 3.0–3.4 and 2.7–3.0, respectively (Bakirci
lowest refrigerant charge (Palm 2007), correlation between
et al. 2011; Cakır et al. 2013). In some studies, the aim of
expansion device and the performance under various
the study was to select the working fluids, which have
charging conditions (Choia and Kim 2004), refrigerant
excellent performance and are environmentally friendly for
mass charge distribution and analysis of proportional gas
moderately high-temperature heat pump (Pan et al. 2011).
mixtures injected to the system (Bjork and Palm 2006;
In another paper, a domestic heating system was designed,
Comaklı et al. 2010) are some of the studies that show the
constructed and tested. The evacuated tubular solar col-
importance of the charge amount in the heat pump systems.
lector was used to achieve higher collector efficiency. The
The overall objective of this study was to develop
effects of evaporation temperature on the heating capacity
energy correlations in air-to-water heat pumps to investi-
and performance of the system were investigated (Caglar
gate the optimal charge amount for different refrigerants
and Yamali 2012). Thermal performance analysis of a
for achieving a better understanding of efficiency of the
direct expansion solar-assisted heat pump and also a sys-
system under various charges and operating conditions by
tem improvements in solar thermal and air source heat
experimental measurement.
pump was carried out by Kong et al. (2011) and Poppi et al.
(2016). The compressor has the key role in the heat pumps
and cooling systems and effects on the gas properties 2 Experimental Setup and Test Procedure
during compression stage. Influence of charge amount on
the performance of a reciprocating compressor (Afshari Experimental setup was designed to measure the perfor-
et al. 2016) and frost formation due to compressor opera- mance of an air-to-water heat pump for applying under
tion and ambient temperature (Tahavvor and Hosseini variable operating conditions. A schematic diagram of the
2014; Tahavvor et al. 2014) were carried out as experi- experimental setup is shown in Fig. 1. In this study, the
mental studies provided in the literature. heat pump unit consists of an open-type reciprocating

123
Iran J Sci Technol Trans Mech Eng (2016) 40:325–335 327

Fig. 1 Schematic diagram of the AWHP experimental setup

compressor, water-cooled copper pipe body-type con- COP, enthalpy, pressure ratio and other related functions
denser, air source copper pipe with aluminum fins evapo- were calculated and compared for all three gases at each
rator, and an expansion device. The condenser was used to gas level.
exchange heat between the refrigerant and water. Air was
selected as a heat source and water acted as a heat sink in
the thermodynamic cycle of constructed heat pump. 3 Calculations
Moreover, used refrigerants were R134a, R404a and R22
(Table 1). Temperatures in the test setup were monitored The heat transfer rate of the condenser and the evaporator
using T-types thermocouples to measure essential points can be calculated from the following equation:
and surface of the condenser, and refrigerant pressures  
Q_ evap ¼ m_ a;e cp;a Tin;a  Tout;a ð1Þ
were also measured at the selected locations with bourdon
 
pressure gauges shown in Fig. 1. In order to reduce heat Q_ cond ¼ m_ w;c cp;w Tout;w  Tin;w ð2Þ
transfer to the environment, all parts of the system
including pipes, water tank, condenser and other compo- In the laboratory heat pump apparatus, three-phase elec-
nents were insulated. The evaporator is installed inside a trical motor has been used as a driving force of the com-
wind tunnel and both speed and temperature of the air pressor. In this state, electrical power consumption of the
inside can be controlled by a fan and an electrical heater. A motor–compressor unit can be calculated as Cakır et al.
volumetric flow meter was installed to measure condenser (2013):
pffiffiffi
cooling water flow rate. Each sensor was calibrated to W_ comp ¼ 3  cos ;  U:I ð3Þ
reduce experimental uncertainties (Karagoz 2002).
The first step of the test procedure was a 2000 g charge where the cos ; is power factor and was taken as 0.8 in the
under the certain condition. Subsequently, the refrigerant calculation.
was added into the heat pump unit in 200–300 g incre- Considering the air fan in the wind tunnel, overall power
ments until the maximum COP was obtained. Charging consumption and the total coefficient of performance
was carried out from 2000 up to 7300 g for R134a, and (COPT) of the heat pump device are as follows:
6000 and 7000 g for R404a and R22 refrigerants, respec- W_ total ¼ W_ comp þ W_ electric fan ð4Þ
tively. Totally, 7 tests were performed in each stage (4 tests
for different condenser water flow rates and 3 tests for Q_ cond
COPT ¼ ð5Þ
different evaporator air temperature). Heating capacity, W_ total

123
328 Iran J Sci Technol Trans Mech Eng (2016) 40:325–335

First law of thermodynamics for condenser can be

defined as:
Q_ cond ¼ m_ r h2  m_ r h3  q_ loss;cond ð10Þ

For compressor, we have:

w_ comp ¼ m_ r h2  m_ r h1 þ e_comp ð11Þ
where the e_comp represents heat and mechanical loss from
the compressor.
Because of being inside of the insulated wind tunnel, we
have negligible heat transfer with ambient in the evaporator
zone and the following relation can be used:
Q_ evap ¼ m_ r h40  m_ r h4 ð12Þ
Fig. 2 Picture of used shell and tube insulated condenser of heat
pump It can be also expressed for expansion valve and heat
exchanger (assuming no heat transfer to ambient),
Heat loss was computed to use in thermodynamics first law respectively:
equations. Proper thickness of asbestos used to reduce heat h3 0 ¼ h4 ð13Þ
loss as shown in Fig. 2, and heat loss from the shell and
tube condenser of the heat pump was calculated by related h3 þ h4 0 ¼ h1 þ h3 0 ð14Þ
relations. By solving last five energy Eqs. (10–14), unknown quan-
Grashof number for our problem can be defined as tities computed and results are available and presented in
Holman (1986): following diagrams (being a nonlinear, this problem has
gBðTas  T1 ÞD3 been solved by Newton–Raphson method using Jacobian
Grd ¼ ð6Þ matrix in MATLAB program and h3 ; h30 ; h4 ; m_ r and e_comp
v2a
has been computed).
Due to the horizontal shape of the condenser, in free
convectional heat transfer state, Nusselt number can be
calculated with a relevant method for a horizontal cylinder 4 Uncertainty Analysis
provided by Holman (1986):

0:518ðGrd pr Þ1=4 Experimental uncertainties can result from instrument

Nu ¼ 0:36 þ h  169 i4=9 ; ð106 \ Gr pr \109 Þ selection, calibration, observation, reading and planning,
1 þ 0:559
Pr and some of them are environmental errors (Bakirci
et al. 2011). In the system, the total uncertainty for
ð7Þ
temperatures measured by copper–constantan thermo-
Heat transfer coefficient is calculated from: couples ±1.2 %, pressures by Bourdon-type 1.6 class
Nu  ka manometers ±1.6 %, electrical power measured by
hc ¼ ð8Þ ampere meter and voltmeter ±3.00 %, air flow rate by
D
anemometer ±1.48 %, water flow rate ±1.2 %, refrig-
We obtain heat loss from the condenser as: erant charging ±0.5 %, and reading values of the table is
2pLðTs  T1 Þ assumed to be ±0.2 % Uncertainties in experimentally
q_ loss;cond ¼   ð9Þ measured quantities can be used in the form of mathe-
ro
ln ri
þ ro1hc matical relationship to calculate considered derived
kas
quantity. Generally, if the output quantity Y is found
First law of thermodynamics can be used for every part of from input quantities X1, X2, …, Xn according to a
heat pump system, to find out unknown quantities. At the function as follows:
intermediate level of the charge amount, because of being Y ¼ FðX1 ; X2 ; . . .Xn Þ ð15Þ
in the two-phase region h3 ; h30 ; h4 are unknown, but fol-
lowing equations are available and used to compute them The combined uncertainty of the output quantity uc(y)
and other desired quantities like refrigerant flow rate and can be calculated via the standard uncertainties of the input
energy loss from the compressor. data u(xi) as:

123
Iran J Sci Technol Trans Mech Eng (2016) 40:325–335 329

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2  2  2 highest point on the chart. After passing through the
oY oY oY
uc ð y Þ ¼ uð x 1 Þ þ uðx2 Þ þ    þ uð x n Þ peak point, additional gas causes a performance reduc-
oX1 oX2 oXn tion and COP value drops in next steps. Furthermore,
ð16Þ these diagrams show that how the different water flow
rates (0.05, 0.06, 0.10 and 0.1314 kg s-1) impact on the
In this case, COP is a function of mass flow rate, water
COP. It is clearly observed that with increasing in flow
input and output temperatures and power consumption of
rate, heat provided from the system increases, which
compressor as shown in Eq. (5).
means a higher performance of the unit.
So the uncertainty of heat pump performance is obtained
There are several factors to evaluate COP, heat
by the following equation (Pan et al. 2011), from which the
capacity and even the optimal charge amount of the
calculated uncertainty for COP (UCOP,HP) was ±5.53 %.
system including inlet temperature, secondary fluids flow

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
s
2
2
2

oCOP oCOP oCOP oCOP 2

UCOP;HP ¼ Um;water
_ þ U t þ U t þ U w_ ð17Þ
om_ cond;water oTcond;water;inlet oTcond;water;outlet oW_

5 Results and Discussion rates, efficiency of heat exchangers, temperature and

consumed power at the compressor, insolation, type of
The allowable refrigerant charge into the heat pump is refrigerant and also type of the used oil in the compres-
connected to the system performance. There are different sor. In most refrigeration and heat pump systems, refrig-
effective parameters to determine refrigerant charging into erant is soluble in the oil, and then, the properties of the
the system: lubricant will change in the system, which influences
efficiency, and even the optimal refrigerant charge of a
• design of the system,
system. In the case that heat pump is in undercharged
• the diameters and lengths of pipes,
condition, the capacity will be at low levels because of
• the refrigerant,
infrequent flow rate of refrigerant. On the other side, in
• size of receivers,
overcharged condition, the COP is also reduced because
• and the expansion device
of an increasing pressure in the system, and consequently,
Refrigerant must provide a range of essential prop- higher compressor power consumption with respect to
erties, the most important ones among them include: low required heat absorption from the condenser does not
gas specific volume, low liquid density of the refrigerant, occur efficiently. Figure 6 shows the effect of gas amount
high vaporization latent heat and a low liquid-phase on the compressor power consumption for all experi-
specific heat (Poggi et al. 2008). Table 2 represents the mented refrigerants.
constant conditions, including temperatures and flow By increasing the gas amount and over charging state,
rates of secondary fluids in condenser and evaporator heating capacity, as shown in Fig. 7, drops also after
applied in the experiments. Figures 3, 4 and 5 show that passing through the peak point obviously. R134a refrig-
how COP value varies as a function of refrigerant charge erant shows a clear reduction, and it is expected for other
amount for R134a, R404a and R22, respectively. Ini- gases to treat like that by more charging with little dif-
tially, COP rises with high gradient until reaching the ference due to gas properties and test conditions.

Table 1 Comparison of various thermodynamic properties of refrigerants

Refrigerant Chemical formula Liquid density Liquid volume Critical Critical Normal ODP GWP
(kg m-3) at 20 C (m3 kg-1) at pressure temperature boiling point
20 C (kPa) (C) (C)

R22 CHClF2 1210.0 0.000826 4990.0 96.15 -40.8 0.05 1700

R134a CH2FCF3 1224.5 0.000816 4060.3 101.08 -26.6 0.0 1430
R404a C2HF5 (44 %)/CH2FCF3 1071.7 0.000938 3732.0 72.07 -46.6 0.0 3800
(4 %)/C2H3F3 (52 %)

123
330 Iran J Sci Technol Trans Mech Eng (2016) 40:325–335

Table 2 Test conditions for secondary fluids in evaporator and condenser

Tcond,in,w Tevap,in,a m_ cond;in;w m_ evap;in;a

11.2 C 19, 20 and 24 C 0.05, 0.06, 0.10 and 0.1314 kg s-1 0.80 kg s-1

R134a R22
3 3

2.5
2.5

2
2

COP
1.5
COP

1.5 0,05 kg/s

0,05 kg/s
1 0,06 kg/s
1 0,06 kg/s
0,10 kg/s
0,10 kg/s 0.5
0.5 0,13 kg/s
0,13 kg/s 0
0 2000 3000 4000 5000 6000 7000
2000 3000 4000 5000 6000 7000 8000 gas amount (gr)
gas amount (gr)
Fig. 5 COP variations with respect to the gas amount and water flow
Fig. 3 COP variations with respect to the gas amount and water flow rate for R22 refrigerant
rate for R134a refrigerant

2800
R404 2700 R134a
2.5
2600 R404a
power consumpon (wa)

2 2500 R22

2400
1.5 2300
COP

2200
0,05 kg/s
1
2100
0,06 kg/s
2000
0.5 0,10 kg/s
1900
0,13 kg/s
0
1800
2000 3000 4000 5000 6000 7000
2000 2500 3000 3500 4000 4500 5000 5500 6000
gas amount (gr) gas amount (gr)

Fig. 4 COP variations according to the gas amount and water flow Fig. 6 Compressor work variation with respect to the charge amount
rate for R404a refrigerant

diagram (Fig. 9) shows refrigerant flow rate variations

Figure 8 presents the heating capacity, COP and power versus gas amount. The flow rate increases with high
consumption variations with respect to the gas amount of slope initially, but step by step decreases with more
R134a refrigerant. The figure shows continuous increasing charge amount which is obviously observed in the case of
power consumption from start to the end of tests. How- R134a and R22 gases. The computed gas flow rate has a
ever, heating capacity and COP value approach the good agreement with the results presented in the literature
maximum point and then reduce swiftly with excessive (Cho et al. 2005).
charging. During experiments, it was determined that in low gas
By solving the first law of thermodynamic equations quantities entire of the thermodynamic cycle is settled in
for all experiments performed, enthalpy in two-phase the superheat region. However, with increasing in gas
region, refrigerant flow rate and compressor losses were amount 3, 3́ and 4 points are dragged to the two-phase
obtained as explained in previous section. Following region. In this region, although temperatures and pressures

123
Iran J Sci Technol Trans Mech Eng (2016) 40:325–335 331

8000 450
super
7000 R134a R22
6000 400
R404a R134a

Enthalpy (kj/kg)
5000
Qh (wa)

R22
350 R404a
4000

3000
300
2000 liquid
1000 250
0
2000 3000 4000 5000 6000 7000 200
gas amount (gr) 2000 3000 4000 5000 6000 7000
gas amount (gr)
Fig. 7 Variation of heating capacity with gas amount for all three
refrigerants Fig. 10 Condenser outlet enthalpy (h3) with respect to the gas
amount
cop Qh power consumpon
4 6000
2500 gr R22
3.5 14
5000 3200 gr
3 12
6600 gr
Qh , w (wa)

4000
2.5 10
Pressure (Bar)
COP

2 3000
8
1.5
2000 6
1
1000 4
0.5
2
0 0
2000 3000 4000 5000 6000 7000 0
gas amount (gr) 100 150 200 250 300 350 400 450 500
enthalpy (kj/kg)
Fig. 8 Heating capacity, COP and power consumption with respect
to the gas amount Fig. 11 Heat pump cycle in P–h diagram for different charge amount

of these points are obtained, there is no possibility to find

out other properties and location of the points in the P–
0.04
h diagram. By solving energy equations, other unknown
quantities like enthalpy and point’s location in two-phase
0.035
region are defined. Figure 10 shows that how the enthalpy
Refrigerant flow rate (kg/s)

of the condenser outlet (point 3) passes through superheat

0.03
and two-phase regions and reaches to the liquid phase
eventually. In the middle district, approximately between
0.025
3000 and 5000 g, condenser outlet enthalpy is placed in
R22
0.02
two-phase region and thermodynamic cycle of the heat
R404a pump is not completed in that condition. However,
0.015
R134a enthalpy comes down rapidly and becomes stable and also
cycle is completed after passing to liquid region subse-
0.01
quently. Figure 11 shows that in very low gas level
2000 3000 4000 5000 6000 7000 (2500 g), entire cycle is in the superheat region, and in this
gas amount (gr) position, performance of the system is less than one. With
additional refrigerant (3200 g), half of the cycle has
Fig. 9 Refrigerant flow rate with respect to the gas amount dragged to the two-phase region, and whole cycle has risen

123
332 Iran J Sci Technol Trans Mech Eng (2016) 40:325–335

3 800 Compressor inlet and discharge pressures can be con-

COP
700 sidered as very significant factors to get required gas

Compressor inlet pressure (kPa)

2.5 amount in a heat pump unit. With the increasing refrigerant
pressure
600
amount, suction line pressure rises and coefficient of per-
2
500 formance increases just until reaching pick point. Heating
capacity is a function of enthalpy and mass flow rate, and
COP

1.5 400
in other hand, power consumption increases due to the
300
1 elevated pressure so there must be an optimal gas charge
200 amount to find highest COP, and finding a relationship
0.5
100
between the charge amount and pressure is one of the aims
of this study. Figures 12, 13 and 14 show relations between
0 0 suction line pressure and COP value, which vary with the
2000 2800 3400 4000 4600 5200 5800 6400 7000
R22- gas amount (gr) gas amount of different gases. This proportion can be
considered to find out needed pressure and a way toward
Fig. 12 Pcomp,in versus COP for R22 refrigerant (Ta,in = 19 C, determining the optimal charge amount in heat pumps.
Tw,in = 11.2, m_ a = 0.8025 kg s-1, m_ w = 0.13 kg s-1) Based on results from the tests, it was found that the full
charge of R134a, R404a and R22 refrigerants of the heat
3 500
COP pump unit was almost 5800–6000, 5200–5400 and 6200 g
compressor inlet pressure (kpa)
2.5 pressure 400
and maximum value obtained for COP was 2.91, 2.40 and
2.78, respectively. In the case of R404a, charging of the
2 system has been performed less than other gases because
300
the compressor power consumption rate has increased
COP

1.5
200 rapidly. The maximum obtained COP values according to
1 the gas amount collected in Table 3, and related com-
100 pressor inlet pressures and charge amount have been listed
0.5
for comparison.
0 0 Compressor working condition varies due to different
2000 2800 3400 4000 4600 5200 5800 6700 parameters including the type of refrigerant and its prop-
R134a - gas amount (gr) erty and also charge amount which effect on inlet and
outlet pressures and temperatures of the compressor.
Fig. 13 Pcomp,in versus COP for R134a refrigerant (Ta,in = 19 C,
Figure 15 shows inlet pressures of the compressor in dif-
Tw,in = 11.2, m_ a = 0.8025 kg s-1, m_ w = 0.13 kg s-1)
ferent conditions, and Fig. 16 shows discharge to inlet
pressure (pressure ratio) variations which decreases with
3 700 increasing in the amount of refrigerant. Among the gases,
COP R404a has less liquid density than others, since it has
600
compressor inlet pressure (kPa)

2.5 presented a higher compressing ratio. Figure 17 represents

pressure
500 the effect of condenser water flow rate on COP for R22
2
400 and R404a refrigerants. For all gases, water flow rate was
COP

1.5 increased in 4 steps and COP increasing has been evalu-

300 ated in various refrigerant levels. The same results were
1 obtained for the evaporator, so that increasing air tem-
200
0.5
perature in the wind tunnel shows a clear increasing of the
100
heat pump performance. It is notable that for R134a
0 0 refrigerant, the temperature of the whole system reduced
2000 2800 3400 4000 4600 5200 5800
due to high gas injection. This refrigerant shows clear
R404a -gas amount (gr) temperature dropping where refrigerant charge is more
Fig. 14 Pcomp,in versus COP for R404a refrigerant (Ta,in = 19 C,
than 5 kg. Even inlet and outlet temperatures of the
Tw,in = 11.2, m_ a = 0.8025 kg s-1, m_ w = 0.13 kg s-1) compressor dropped with an increase in the charge
amount. Figure 18 shows a decrease in the inlet temper-
ature of the all experimented gases in the overcharged
up because of increasing pressure in next steps. In the next condition, and in Fig. 19, inlet temperature and outlet
stage (6600 g), charging is enough to complete the cycle temperature of the R404a refrigerant are compared with
and desired procedure is established. each other in various water mass flow rates.

123
Iran J Sci Technol Trans Mech Eng (2016) 40:325–335 333

Table 3 Optimum COP due to gas amount and compressor inlet pressure for all gases
Refrigerant m_ w = 0.05 kg s-1 m_ w = 0.06 kg s-1 m_ w = 0.10 kg s-1 m_ w = 0.13 kg s-1
COP P1 (kPa) Charge (g) COP P1 (kPa) Charge (g) COP P1 (kPa) Charge (g) COP P1 (kPa) Charge (g)

Ta,in = 19 C, Tw,in = 11.2, m_ a = 0.8025 kg s-1

R22 2.22 650.27 6200 2.35 650.27 6200 2.55 649.46 6200 2.78 646.72 6200
R134a 2.27 431.61 6000 2.49 417.13 5800 2.74 410.92 5800 2.75 406.79 5800
R404a 1.63 426.09 5200 2.03 432.99 5200 2.19 457.12 5200 2.40 474.35 5200
Refrigerant Ta,in = 19 C Ta,in = 20 C Ta,in = 24 C
COP P1 (kPa) Charge (g) COP P1 (kPa) Charge (g) COP P1 (kPa) Charge (g)

m_ w = 0.10 kg s-1, Tw,in = 11.2, m_ a = 0.8025 kg s-1

R22 2.54 653.62 6200 2.58 681.20 6200 2.74 698.43 6200
R134a 2.27 431.61 6000 2.65 422.64 5800 2.91 446.78 5800
R404a 2.27 455.05 5200 2.32 488.14 5400 2.36 491.59 5400

800 R404a- 4800gr

R134a 2.5 R404a- 5000gr

700 2.4 R404a -5200gr
R404a R22-4800gr
600 2.3 R22-5000gr
R22 R22-5200gr
pressure (kPa)

500 2.2
2.1
400
COP

2
300
1.9
200 1.8
100 1.7
1.6
0
2000 3000 4000 5000 6000 7000 1.5
gas amount (gr) 0.04 0.06 0.08 0.1 0.12 0.14
water flow rate (kg/s)
Fig. 15 Variation of Pcomp,in with charge amount
Fig. 17 COP variation with respect to condenser water flow rate for
R22 and R404a refrigerants (charge amount 4800, 5000 and 5200 g)
7
R134a
R404a
In the heat pump system, heat transfer rate increases as a
6
R22 consequence of the difference between the temperature of
secondary fluid and condensing temperature. Additionally,
5
pressure rao

both condensing and evaporating pressures change by

4 secondary fluid temperatures. In this type of heat pumps
(AWHP), air temperature affects the evaporating pressure
3 and also condensing pressure is a function of cooling water
temperature.
2 In general, the charge amount affects the condensing
and evaporating pressures of the cycle, and with increasing
1 in the refrigerant charge amount, condensing and evapo-
2000 3000 4000 5000 6000 7000
rating pressures increase as well as density of the gas inside
gas amount (gr)
the system. Refrigerant flow rate is directly related to the
Fig. 16 Pressure ratio of the compressor versus gas amount for all gas amount and increases with the charging. Figure 20
gases shows the variation of compression ratio and mass flow

123
334 Iran J Sci Technol Trans Mech Eng (2016) 40:325–335

22 compression rao
3.5 0.035
3.3 gas flow rate
compressor inlet temperature (°C)

20 0.03
3.1

compression rao
0.025

gas flow rate (kg/s)

18 2.9
2.7 0.02
16
2.5
0.015
2.3
14
R134a
2.1 0.01
R404a
12 1.9
R22 0.005
1.7
10
1.5 0
2500 3000 3500 4000 4500 5000 5500 6000 6500 7000
gas amount (gr) 3000 3800 4600 5400 6200 7000
gas amount (gr)
Fig. 18 Variations of compressor inlet temperature with gas amount
(Ta,in = 19 C, Tw,in = 11.2, m_ a = 0.8025 kg s-1, Fig. 20 Compression ratio and gas flow rate variations according to
m_ w = 0.13 kg s-1) the charge amount

90 35 15
R22

85 33
13 R404 a
31
Outlet temperature (°C)

R134a
80
pressure (bar)

29 11
Inlet temperature (°C)

T,comp,out- 0.05kg/s
75 0.06 kg/s 27
0.1 kg/s 9
70 0.13 kg/s 25
T,comp,in- 0.05kg/s
0.06 kg/s 7
65 23
0.10 kg/s
0.13 kg/s
21
60 5
19
55 17 3
200 250 300 350 400 450
50 15
2000 3000 4000 5000 6000 enthalpy (kj/kg)
Gas amount (gr)
Fig. 21 Charge effect (5400, 5600 and 5800 g) on the heat pump
Fig. 19 Compressor inlet and outlet temperature with respect to the cycle in the P–h diagram (Ta,in = 19 C, Tw,in = 11.2,
gas amount in different water flow rates (kg s-1) at the condenser m_ a = 0.8025 kg s-1, m_ w = 0.13 kg s-1)
(purple and yellow lines indicate inlet and outlet results, respectively)

evaporating pressure. Generally, variation in the system

rate of refrigerant with charge amount for R134a refriger-
pressure changes the evaporator and condenser enthalpies.
ant. Although the system pressure increases in both the
condenser and the evaporator, pressure ratio decreases
gradually with the additional charge and mass flow rate
6 Conclusion
increases as shown in this figure.
Figure 21 presents the pressure–enthalpy cycle of three
In this study, the effects of refrigerant charge on the COP
refrigerants in constant test conditions at the condenser and
of the air-to-water heat pump were investigated using
evaporator. This figure shows that each gas works in
R404a, R143a and R22 refrigerants under various working
individual cycle, which is proportional to the pressure and
conditions. The main findings of this study can be sum-
temperature of the system provided by refrigerant type and
marized as follows;
gas amount.
Obvious changes in condensing and evaporating pressures 1. Possibility to find out the optimal charge amount in a
related to the gas amount are shown in this figure. Compar- heat pump system has been presented for different
ison of the results for all cases shows the increasing charge refrigerants with different properties with respect to the
amount influence on the condensing pressure more than suction pressure in the compressor of the system.

123
Iran J Sci Technol Trans Mech Eng (2016) 40:325–335 335

2. It was determined that the maximum COP of the Cho H, Ryu C, Kim Y, Kim H (2005) Effects of refrigerant charge
system strongly depends on the charge amount and amount on the performance of a transcritical CO2 heat pump. Int
J Refrig 28:1266–1273
refrigerant type. On the other side, performance of the Choi H, Cho H, Choi J (2012) Refrigerant amount detection algorithm for
system decreases in overcharge condition. a ground source heat pump unit. Int J Renew Energy 42:111–117
3. This study revealed that in the low gas levels whole Choia J, Kim Y (2004) Influence of the expansion device on the
cycle is in the superheat region in the P–h diagram. performance of a heat pump using R407C under a range of
charging conditions. Int J Refrig 27:378–384
With additional charge, the cycle passes through two- Comaklı O, Bayramoglu M, Kaygusuz K (1996) A thermodynamic
phase region at first and then reaches rapidly liquid model of a solar assisted heat pump system with energy storage.
region which the cycle will be finally completed in last Int J Pergamon 56:485–492
stage. Comaklı K, Yuksel B, Comaklı O (2004) Evaluation of energy and
exergy losses in district heating network. Int J Appl Therm Eng
4. The effects of gas type and charging level on the 24:1009–1017
pressure and temperature of the compressor were Comaklı O, Comakli K, Ozdemir N, Yilmaz M (2010) Analysis of
characterized in suction and discharge lines. heat pumps with zeotropic refrigerant mixtures by Taguchi
5. Obtained results show that compression ratio is very method. J Therm Sci Technol 30(2):113–122
Corberan J, Martinez I, Gonzalvez J (2008) Charge optimization
variable parameter and strongly depends on gas study of a reversible water-to-water propane heat pump. Int J
properties and charge level. Refrig 31:716–726
6. It was specified that refrigerant property is a significant Corberan J, Martinez I, Martinez-Ballester S, Gonzalvez-Macia J,
parameter for working condition of a heat pump; Royo-Pastor R (2011) Influence of the source and sink temper-
atures on the optimal refrigerant charge of a water-to-water heat
especially, gas density has a major role in temperature pump. Int J Refrig 34:881–892
and pressure of the system and also pressure ratio in Fernando P, Palm B, Lundqvist P, Granryd E (2004) Propane heat
the condenser. Some refrigerants cause an increase in pump with low refrigerant charge: design and laboratory tests.
temperature and pressure of compressor that leads us to Int J Refrig 27:761–773
Holman JP (1986) Heat transfer, 6th edn. ISBN: 0-07-029620-0
select a proper compressor and even lubrication oil in Karagoz S (2002) Utilization of refrigerant mixtures in heat pumps,
heat pumps. MSc thesis
Kim D, Park H, Kim M (2014) The effect of the refrigerant charge
The scientific method and obtained results presented in amount on single and cascade cycle heat pump systems. Int J
this study can be used for industrial process improvement Refrig 40:254–268
of heat pumps and refrigerants. Kong X, Zhang D, Li Y, Yang Q (2011) Thermal performance
analysis of a direct-expansion solar-assisted heat pump water
heater. Int J Energy 36:6830–6838
Acknowledgments This project was supported by the Research
Palm B (2007) Refrigeration systems with minimum charge of
Project Foundation of the Ataturk University (Project No. BAP-2013/
refrigerant. Int J Appl Therm Eng 27:1693–1701
105). The authors gratefully acknowledge the support of this study.
Pan L, Wang H, Chen Q, Chen C (2011) Theoretical and experimental
study on several refrigerants of moderately high temperature
heat pump. Int J Appl Therm Eng 31:1886–1893
Poggi F, Macchi-Tejeda H, Leducq D, Bontemps A (2008) Refrig-
References erant charge in refrigerating systems and strategies of charge
reduction. Int J Refrig 31:353–370
Afshari F, Comakli O, Adiguzel N, Ghasemi Zavaragh H (2016) Poppi S et al (2016) Analysis of system improvements in solar
Influence of refrigerant properties and charge amount on thermal and air source heat pump combisystems. Appl Energy
performance of reciprocating, compressor in air source heat 173:606–623
pump. J Energy Eng. doi:10.1061/(ASCE)EY.1943-7897.0000377, Tahavvor AR, Hosseini S (2014) The effects of humidity, compressor
04016025 operation time and ambient temperature on frost formation in an
Bakirci K, Ozyurt O, Comakli K, Comakli O (2011) Energy analysis industrial fridge, iranian journal of science and technology.
of a solar-ground source heat pump system with vertical closed- Trans Mech Eng 39:197–203
loop for heating applications. Int J Energy 36:3224–3232 Tahavvor AR, Zakeri E, Moezi SA (2014) Modeling of frost growth
Bjork E, Palm B (2006) Refrigerant mass charge distribution in a on a horizontal circular cylinder under natural convection using
domestic refrigerator. Part I: transient conditions. Int J Appl fractal geometry analysis. Iran J Sci Technol Trans Mech Eng
Therm Eng 26:829–837 38:1–8
Caglar A, Yamali C (2012) Performance analysis of a solar-assisted Yan G, Bai T, Yu J (2016) Energy and exergy efficiency analysis of
heat pump with an evacuated tubular collector for domestic solar driven ejector–compressor heat pump cycle. Sol Energy
heating. Int J Energy Build 54:22–28 125:243–255
Cakır U, Comaklı K, Comaklı O, Karsli S (2013) An experimental Zhang D, Wu QB, Li JP, Kong XQ (2014) Effects of refrigerant
exergetic comparison of four different heat pump systems charge and structural parameters on the performance of a direct-
working at same conditions: as air to air, air to water, water to expansion solar-assisted heat pump system. Int J Appl Therm
water and water to air. Int J Energy 58:210–219 Eng 73:520–526

123