Content uploaded by Steffen Feja
Author content
All content in this area was uploaded by Steffen Feja on Dec 20, 2017
Content may be subject to copyright.
Measurement of electrical properties of refrigerants and
refrigeranteoil mixtures
S. Feja*
ILK Dresden gGmbH, Bertolt-Brecht-Allee 20, Dresden 01309, Germany
article info
Article history:
Received 4 November 2011
Received in revised form
7 March 2012
Accepted 11 March 2012
Available online 21 March 2012
Keywords:
Air conditioning
Dielectric constant
Dielectric property
Electric resistance
R134a
R744
R152a
R1234yf
abstract
The paper deals with the description of a new pressure proof device developed at ILK
Dresden for measuring electrical properties of refrigerants, oils and refrigeranteoil
mixtures under pressure. The determination can be performed based on DIN EN 60247.
This standard is valid for the determination of the relative permittivity, the dielectric
dissipation factor and the direct current (DC) resistivity of isolating fluids at atmospheric
pressure. The measurements can be made in a temperature range from 30 Cto90Cat
pressures up to 80 bar under dry atmospheric conditions. The sensitivity of the method as
well as its considerable flexibility meets the requirements of the industrial and commercial
refrigeration. Challenges regarding the development and performance of the measuring
cell as well as some preliminary experimental results are presented. The focus of the
experiments is on environmentally friendly refrigerants (GWP <150) such as R744, R1234yf
and R152a, and their mixtures with lubricants.
ª2012 Elsevier Ltd and IIR. All rights reserved.
Mesures des proprie
´te
´se
´lectriques des frigorige
`nes et des
me
´langes de frigorige
`ne / huile
Mots cle
´s:Conditionnement d’air ; Constante die
´lectrique ; Proprie
´te
´die
´lectrique ; Re
´sistance e
´lectrique ; R134a ; R744 ; R152a ; R1234yf
1. Introduction
Due to the increasing applications of hermetic and semi-
hermetic compressors in refrigeration the question of the
electrical properties of the refrigeranteoil mixtures used is
becoming an important issue. The MAC-systems in electro
and hybrid vehicles, whose electrical systems will be designed
for voltages up to 500 V, are only one example for this interest.
The knowledge about the specific resistivity of the working
fluid mixtures and their application in electrical refrigeration
machines is becoming more and more important. Beyond that
the knowledge on the dielectric permittivity leads to
* Tel.: þ49 351 4081 767; fax: þ49 351 4081 755.
E-mail address: Steffen.Feja@ilkdresden.de.
www.iifiir.org
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/ijrefrig
international journal of refrigeration 35 (2012) 1367e1371
0140-7007/$ esee front matter ª2012 Elsevier Ltd and IIR. All rights reserved.
doi:10.1016/j.ijrefrig.2012.03.011
fundamental statements about the behaviour of molecules in
an electrical field, which could be used for energetic optimi-
zation of air conditioning systems. Although the determina-
tion of the permittivity (dielectric constant, ε
x
) of oils has been
performed at the ILK Dresden since 1965, very little has been
published on this topic (e.g. Pa
¨tz et al., 1968). The analysis of the
dielectric constant and the DC resistivity of refrigerants and
refrigeranteoil mixtures in the relevant temperature interval
between 30 C and þ90 C and at pressures up to 100 bar is
quite challenging. Only a few research groups around the world
have been working on this subject (Barao et al., 1995,1996,1998;
Baustian et al., 1986;Bo
¨hmer and Loid, 1988;Fellows et al., 1991;
Hwang et al., 2008;Meurer et al., 2001;Ribeiro & Nieto de Castro,
2009;Tanaka et al., 1999) mostly using pure substances, but no
refrigeranteoil systems.
Actually, no measuring standards for the determination of
the permittivity and electrical resistivity of refrigeranteoil
systems are defined. Therefore, the investigations are per-
formed with a novel test setup, which is based on DIN EN
60247 (2005),ASTM D 924 (2008) and ASTM D 1169 (2011) set
for the determination of electrical properties of transformer
oils. The capacity and the resistivity of a cylinder gap appa-
ratus without protective ring have been measured directly
filled with air (10
2
mbar), calibrations substances and test
fluids in the gap. The permittivity and the specific resistivity of
the test fluids are calculated from these measurements
according DIN EN 60247 (2005).
2. Experimental methodology
With respect to the nature of refrigerants and refrigerant
mixtures the measurements require a pressure-tight cell,
which allows investigation in a homogeneous electrical field
necessitating homogeneous refrigeranteoil mixtures.
However, in addition to these requirements the enormous
sensitivity of high-resistance and permittivity measurements
are challenging. Especially the permittivity is very sensitive to
external influences (e.g. partial capacitances, impurities,
evolving pressure). Furthermore, the measurement setup has
to cover a high measurement range with respect to the pure
working fluids, i.e. oils and refrigerants. DC resistivities
ranging from 10
5
to 10
15
Um have to be determinable at
constant current. On the other hand, the examination of the
capacitance used to calculate the permittivity requires alter-
nating current.
Thus, the redesign of the measurement cell had to be
performed with respect to the required experimental param-
eters including:
- thermal control (30 Cto90C) of the test cell and both
electrodes
- pressure proof (up to 80 bar refrigerant pressure).
Fig. 1 shows the electrode setup of the test cell. The gap
between the two electrodes is 2 mm. The outer temperature
and pressure sensor (Kulite HKL/T 375 M) was screwed in the
visible NPT screwing in the outer electrode. Due to the high
voltage of 500 V required for the resistivity measurement as
well as the partial capacitance generated by the sensor this
part of the cell had to be reconstructed. Therefore, the pres-
sure sensor was fixed between the storage container and the
test cell and electrically decoupled from the outer electrode by
a pressure-tight synthetic hose.
The temperature of the outer electrode is defined by the
temperature of the surrounding thermostat bath. The test cell
is embedded in a silicon bath distinguished by a low dielectric
constant and an electrical conductivity below 10
12
Sm
1
.As
temperature sensor a Pt100 is inserted into the inner elec-
trode. A Precision Component Analyzer 6425 from Wayne Kerr
is used as the capacitance detection unit. Additionally a Ter-
raOhmmeter TO-3 from FISCHER ELEKTRONIK is utilized for
detection of the DC resistivity.
Furthermore a storage container which acts as a sample
reservoir for pre-conditioning of the refrigeranteoil mixtures
is connected to the test cell (Fig. 2).
The capacitance measurement is performed at 1 kHz AC
and 1 V. To control the values of the capacitance measure-
ment and to get further information on the dependence of the
dissipation factor from the properties of the liquid the dissi-
pation factor and the AC resistance is measured too. After the
AC measurement the cell is short-circuited for 1 min. After
this the electrodes are connected with the TerraOhmmeter.
After another minute the zero current is determined. This
zero current should be 1000 times less than the measuring
current to avoid an influence on the DC measurement. Next
the DC voltage is applied to the system. The DC resistance is
measured at 500 V given a load of 250 V/mm according to DIN
EN 60247 (2005). If the fluids exhibit a resistivity lower than
10
6
Um the voltage is set to 1 V instead of 500 V. This is done
because the current becomes too high and therefore is not
within the measurement range of the TerraOhmmeter used.
The voltage is set to 1 V so that the measurement range is as
wide as possible. 1 min after turning on the voltage the
resistance is measured and after a further minute the current
is measured to check the influence of the zero current.
The temperature range of the measurements is set above
the ambient temperature because of condensing water at the
electrode top surface. The resulting partial capacitances affect
Nomenclature
ε
x
dielectric constant, permittivity
Ttemperature (K)
wtemperature (C)
xconcentration (wt%)
C
a
idle capacity (empty cell) F
C
g
correction capacity F
C
e
electrode constant F
C
n
capacity with calibration liquid F
C
x
capacity with measuring liquid F
ε
n
dielectric constant calibration liquid
rspecific resistivity Um
Uvoltage V
Icurrent A
Kcell constant m
international journal of refrigeration 35 (2012) 1367e13711368
the determined permittivity of the sample. In the worst case,
the condensed film connects the inner and the outer electrode
which prevents a precise measurement of the electrical prop-
erties. Efforts to overcome this shortcoming are being made.
3. Experimental results
3.1. Calibration
According to DIN EN 60247 the test cell has to be calibrated at
room temperature in empty and liquid-filled state (calibration
substance). The idle capacity is marginally affected by
temperature in the range of 30 Cto90C. The capacity of the
empty cell without protective ring is divided into the parts
correction capacity (C
g
) and electrode constant (C
e
)(DIN EN
60247 (2005)) (Eq. (1)).
Ca¼CeþCg(1)
The terms C
g
and C
e
can be determined using the calibration
substance following Eqs. (1) and (2).
Ce¼CnCa
εn1(2)
Although the capacity of the cell filled with calibration
substance is temperature dependent, the terms C
e
and C
g
are
temperature independent. The permittivity ε
x
of an unknown
liquid is now determined using C
x
from the capacitance
measurement according to Eq. (3).
εx¼CxCg
Ce
(3)
For determining the specific DC resistivity the resistance of
the filled cell is determined at constant voltage. In DIN EN
60247 the specific DC resistivity is calculated according to
Eq. (4).
r¼KU
Iwith K¼0:113$capacity of the empty cell (4)
Using the idle capacity C
a
in this equation (Eq. (4)) is only
valid for cells constructed with a protective ring. After
comprehensive study using our cell, the value of the electrode
constant C
e
must be used as the capacity of the empty cell in
Eq. (4) for cells without a protective ring. Due to the wide range
of DC resistivities of insulating fluids from 10
5
to 10
15
Um this
misinterpretation of the DIN EN 60247 (2005) is not crucial
using the correct oil for hermetic compressors, but should be
considered comparing the results from different suppliers.
The test cell was calibrated after each cleaning process
with n-Heptane as calibration liquid at room temperature.
The cell was also calibrated with more polar fluids such as
Acetone and Fluorobenzene. These calibrations gave the same
results as the calibration with n-Heptane, but it was very time
Fig. 1 ea: Inner (left) and outer electrode (right). b: Assembled measurement cell.
Fig. 2 eBlock diagram of the test setup to determine electrical properties of refrigeranteoil mixtures.
international journal of refrigeration 35 (2012) 1367e1371 1369
consuming to remove the water content from the substances.
In fact the water content influences the measurement of the
dielectric constant and so a calibration with these fluids is too
complicated for daily measurements under normal laboratory
conditions.
3.2. Measurements of refrigerants, oils and
refrigeranteoil mixtures
First the pure refrigerants R134a, R1234yf, R152a and R744
were measured (Fig. 3). These refrigerants are the most
popular candidates for use in electro-mobile air conditioning
systems. The results of the permittivity measurement of
R134a were in agreement with the results given in Barao et al.
(1995),Meurer et al. (2001),Barao et al. (1996) and Fellows et al.
(1991). Also the value of the permittivity of R152a is compa-
rable to the results presented in Barao et al. (1998) and Fellows
et al. (1991) (Fig. 4). The permittivity of all four refrigerants
decreases with increasing temperature whereas the resistivity
doesn’t show a consistent temperature dependence. With
increasing temperature the resistivity decreases (R134a),
increases (R1234yf), or remains almost constant (R152a).
Comparing permittivity and resistivity values, R152a
exhibits the highest permittivity and the lowest resistivity in
comparison to the other refrigerants. The resistivity of R134a
and of R1234yf is nearly the same. The resistivity of liquid and
supercritical R744 is higher than 10
14
Um (data not shown).
The measurements were carried out at the saturation pres-
sure of the fluids.
After measuring the pure refrigerants, refrigeranteoil
mixtures were measured using at least one commercial oil
(Fig. 4). Interestingly, the permittivity of a mixture containing
nearly 75 wt% oil does not differ much from the permittivity of
the pure oil whereas the addition of 50 wt% oil to the refrig-
erant significantly decreases the permittivity compared to the
pure refrigerant sample. The same could be observed in case
of the resistivity measurements. No linear correlation has
been observed between the concentration and the electrical
properties of the pure fluids.
Before and after the measurements the water contents of
the oils were measured to check the influence of water on the
electrical properties.
4. Conclusion and outlook
A device to determine electrical properties of refrigeranteoil
mixtures under pressure has been developed. The test device
is based on the existing method of measuring the electrical
properties of isolating fluids under atmospheric pressure, as
described in the literature. The electrical properties direct
current resistivity and permittivity of refrigerants, oils and
refrigeranteoil mixtures were determined in a temperature
range of 20 Ce90 C. Further efforts are being made to design
Fig. 3 ePermittivity and resistivity of R744, R134a, R1234yf and R152a.
Fig. 4 ePermittivity and resistivity of R152a mixed with the POE oil ISO VG 80.
international journal of refrigeration 35 (2012) 1367e13711370
a measuring cell which allows the determination of electrical
properties of refrigeranteoil mixtures below ambient
temperature. Apart from that, the influence of the water
content and the impurities of oil and-/or refrigerant on the
electrical properties (possibly monitored by sensors) will be
subject of further studies.
Acknowledgments
This work was supported by the grant of the Bundesministe-
rium fu
¨r Wirtschaft und Technologie “FuE-Fo
¨rderung
gemeinnu
¨tziger externer Industrieforschungseinrichtungen
in Ostdeutschland eInnovationskompetenz (INNO-KOM-
Ost)” Modul Vorlaufforschung VF090028. The author wishes to
thank Jan Hegewald (Hegewald & Peschke, Mebund Pru
¨f-
technik GmbH, Nossen) for establishing and testing the
apparatus and Steven Rhode (TU Dresden) for his help with
the measurements. Special thanks go to Sven Heinrich (H. -P.
FISCHER ELEKTRONIK GmbH & Co.) for supporting us with the
development of the measuring cell.
references
ASTM D 924, 2008. Standard Test Method for Dissipation Factor
(Or Power Factor) and Relative Permittivity (Dielectric
Constant) of Electrical Insulating Liquids. ASTM International.
ASTM D 1169, 2011. Standard Test Method for Specific Resistance
(Resistivity) of Electrical Insulating Liquids. ASTM
International.
Barao, T., Nieta de Castro, C.A., Mardolcar, U.V., Okambawa, R.,
St-Arnaud, J.M., 1995. Dieelectric constant, dielectric virial
coefficient and dipole moments of 1,1,1,2-Tetrafluorethane. J.
Chem. Eng. Data 40, 1242e1248.
Barao, M.T., Nieto de Castro, C.A., Mardolcar, U.V., 1996. The
dielectric constant of liquid HFC 134a and HCFC 142b. Int. J.
Thermophys. 17, 573.
Barao, M.T., Mardolcar, U.V., Nieto de Castro, C.A., 1998. Dielectric
constant and dipole moments of 1,1,1-Trifluoro-2,2-
Dichloroethane (HCFC 123) and 1,1-Difluoroethane (HFC 152a)
in the liquid phase. Fluid Phase Equil. 1753, 150e151.
Baustian, J.J., Pate, M.B., Bergles, A.E., 1986. Properties of oil-
refrigerant liquid mixtures with applications to oil
concentration measurement: part II-electrical and optical
properties. ASHRAE Trans. 92 (Pt.1), 74e92.
Bo
¨hmer, R., Loid, A., 1988. Dielectric properties of condensed
fluoromethanes and fluoromethane mixtures. J. Chem. Phys.
89 (8), 4981e4986.
DIN EN 60247, 2005. Isolierflu
¨ssigkeiten eMessung der
Permittivita
¨tszahl, des dielektrischen Verlustfaktors (tan d) und
des spezifischen Gleichstrom-Widerstandes. DIN EN. http://
www.beuth.de/cn/J-650DE375B16CD477738B952469BEEDA9.2/
d29ya2Zsb3duYW1lPWV4YUJhc2ljU2VhcmNoJnJlZj10cGwta
G9tZSZsYW5ndWFnZWlkPWRl.html.
Fellows, B.R., Richard, R.G., Shankland, I.R., 1991. Electrical
characterization of alternate refrigerants. In: Actes Congr. Int.
Froid, 18th Seint-Hyacinthe, Que, vol. 2, pp. 398e402.
Hwang, Y., Radermacher, R., Hirata, T., 2008. Oil mass fraction
measurement of CO
2
/PAG mixture. Int. J. Refrigeration 31,
256e261.
Meurer, C., Pietsch, G., Haacke, M., 2001. Electrical properties
of CFC- and HCFC-substitutes. Int. J. Refrigeration 24,
171e175.
Pa
¨tz, G., Ha
¨ntzschel, H., Neubert, J., 1968. Messung elektrischer
Eigenschaften von Ka
¨ltemaschineno
¨len. Fachbericht Nr. 97/
68. ILK Dresden.
Ribeiro, A.P.C., Nieto de Castro, C.A., 2009. Dielectric properties of
liquid refrigerants: facts and trends. In: IIR 3rd Conference on
Thermophysical Properties and Transfer Processes of
Refrigeration, Boulder, CO, paper No 108.
Tanaka, Y., Matsuo, S., Sotani, T., Kondo, T., Matsuo, T., 1999.
Relative permittivity and resistivity of liquid HFC
refrigerants under high pressure. Int. J. Thermophys. 20 (1),
107e117.
international journal of refrigeration 35 (2012) 1367e1371 1371
