Int. J. Metrol. Qual. Eng.

9, 7 (2018) International Journal of

© G.S. Sariyerli et al., published by EDP Sciences, 2018 Metrology and Quality Engineering
https://doi.org/10.1051/ijmqe/2018004
Available online at:
www.metrology-journal.org

RESEARCH ARTICLE

Comparison tests for the determination of the viscosity values

of reference liquids by capillary viscometers and stabinger
viscometer SVM 3001
Gokce Sevim Sariyerli*, Orhan Sakarya, and Umit Yuksel Akcadag
The Scientific and Technical Research Council of Turkey, National Metrology Institute (TUBITAK UME), Gebze, Turkey

Received: 27 December 2017 / Accepted: 13 March 2018

Abstract. The present study was realized for measuring viscosities of reference liquids using capillary
viscometers and Stabinger viscometer SVM 3001 with viscosity interval between 1 mm2/s and 5000 mm2/s with
temperatures from 20 °C to 80 °C. Based on our measurement with various liquids, we determine the viscosity
values and compare both of the results. The aim of this study to evaluate the results of the primary level viscosity
measurement system and stabinger viscometer and to compare the measurement results due to the providing
traceability of Stabinger viscometer by TUBITAK UME. An increasing number of national metrology institutes
and accredited laboratories provide viscometer calibration with reference liquids in a wide viscosity range. It is a
common practice to use the viscosity of water as the metrological basic of viscometry. The national standard of
viscosity provided by TUBITAK UME consists of a set of ubbelohde viscometers covering the measuring range
of kinematic viscosities from about 0.5 mm2/s to 100 000 mm2/s. At the low viscosity, long
capillary
viscometers are used as primary standards which are directly calibrated water.
Keywords: viscosity / capillary / stabinger viscometer / traceability / metrology

1 Introduction – directly proportional to the pressure difference (DP)

between the ends of the tube;
Viscosity is the ratio of the shearing stress (F/A) to the – inversely proportional to the length (ℓ) of the tube;
velocity gradient (Dvx/Dz or dvx/dz) in a fluid. The more – inversely proportional to the viscosity (h) of the fluid;
usual form of this relationship, called Newton’s equation, – proportional to the fourth power of the radius (r4) of the
states that the resulting shear of a fluid is directly tube.
proportional to the force applied and inversely proportional Viscosity measurements are important in many fields of
to its viscosity. The similarity to Newton’s second law of both industry and research. Many manufacturers now
motion (F = m.a) should be apparent. regard viscometers as a crucial part of their research,
Kinematic viscosity is a measure of the resistive flow of development, and process control programs. They know
a fluid under the influence of gravity. It is frequently that viscosity measurements are often the quickest, most
measured using a device called a capillary viscometer. accurate and most reliable way to analyze some of the most
When two fluids of equal volume are placed in identical important factors affecting product performance.
capillary viscometers and allowed to flow under the Viscosity is a technological quantity which is related
influence of gravity, a viscous fluid takes longer than a with the flow of matter. The viscosity measurements find
less viscous fluid to flow through the tube. great importance in rheology. A frequent reason for the
Mathematical expression describing the flow of fluids in measurement of rheological properties can be found in the
circular tubes was determined as below: area of quality control, where raw materials must be
consistent from batch to batch. For this purpose, flow
pDPr4 behavior is an indirect measure of product consistency and
qm ¼ ;
8hl quality.
Another reason for making flow behavior studies is that
a direct assessment of processability can be obtained. For
* Corresponding author: [email protected] example, a high viscosity liquid requires more power to

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
2 G.S. Sariyerli et al.: Int. J. Metrol. Qual. Eng. 9, 7 (2018)

pump than a low viscosity one. Knowing its rheological Table 1. APS3 Reference liquid (Method 2).
behavior, therefore, is useful when designing pumping and
piping systems. Temperature (°C) Kinematic U0 (n) (%)
It has been suggested that rheology is the most sensitive viscosity (mm2/s)
method for material characterization because flow behav-
20 4.537 2
ior is responsive to properties such as molecular weight and
molecular weight distribution. This relationship is useful in 25 4.006 2
polymer synthesis, for example, because it allows relative 40 2.880 2
differences to be seen without making molecular weight 60 2.023 2
measurements. 80 1.511 2
The instruments for viscosity measurements are
designed to determine “a fluid’s resistance to flow,” a fluid
property defined above as viscosity. The fluid flow in a
given instrument geometry defines the strain rates, and the Table 2. APN7.5 Reference liquid (Method 2).
corresponding stresses are the measure of resistance to flow.
If strain rate or stress is set and controlled, then the other Temperature (°C) Kinematic U0 (n) (%)
one will, everything else being the same, depend on the fluid viscosity (mm2/s)
viscosity. If the flow is simple (one dimensional, if possible) 20 12.42 2
such that the strain rate and stress can be determined 25 10.52 2
accurately from the measured quantities, the absolute
40 6.785 2
dynamic viscosity can be determined; otherwise, the
relative viscosity will be established. For example, the 60 4.246 2
fluid flow can be set by dragging fluid with a sliding or 80 2.917 2
rotating surface, falling body through the fluid, or by
forcing the fluid (by external pressure or gravity) to flow
through a fixed geometry, such as a capillary tube, annulus, Table 3. APN26 Reference liquid (Method 2).
a slit (between two parallel plates), or orifice. The
corresponding resistance to flow is measured as the Temperature (°C) Kinematic U0 (n) (%)
boundary force or torque, or pressure drop. The flow rate viscosity (mm2/s)
or efflux time represents the fluid flow for a set flow
resistance, like pressure drop or gravity force [1]. 20 61.70 2
The basic principle of all viscometers is to provide as 25 49.65 2
simple flow kinematics as possible such as the utilization of 40 27.78 2
one-dimensional flows in order to characterize and 60 14.89 2
determine the shear strain rate as accurately, easily and
80 9.08 2
as independently of the fluid type as possible. The
resistance to such flow is measured, and thereby the
shearing stress is determined. The shear viscosity is then
easily found as the ratio between the shearing stress and the 2 Methods
corresponding shear strain rate [1].
Glass capillary viscometers are very simple and 2.1 Reference liquids and viscometers
inexpensive. Their geometry resembles a U-tube with at
least two reservoir bulbs connected to a capillary tube Reference liquids were obtained Paragon Scientific Ltd.,
passage with inner diameter D. The fluid is drawn up into UK. The names of the liquids are APS3, APN7.5, APN 415
one bulb reservoir of known volume, V0, between etched and APN26.
marks. The efflux time, Dt, is measured for that volume to The comparison tests were realized at 20 °C, 25 °C,
flow through the capillary under gravity. Various kinds of 40 °C, 60 °C and 80 °C using the reference liquids (APS3,
commercial glass capillary viscometers, like Cannon- APN26, APN7.5 and APN 415) by both Ubbelohde
Fenske type or similar, can be purchased from scientific viscometers and Stabinger viscometer SVM 3001. For
and/or supply stores. They are the modified original Method 2, the results of the reference liquids (APS 3, APN
Ostwald viscometer design in order to minimize certain 26, APN 7.5 and APN 415) were given in Table 1, Table 2,
undesirable effects, to increase the viscosity range, or to Table 3 and Table 4.
meet specific requirements of the tested fluids, like opacity, Method 1, the two Ubbelohde viscometers with type 0B
etc. Glass capillary viscometers are often used for low- (K = 0.005 mm2/s2) were used for the determination of
viscosity fluids [1]. kinematic viscosity values of APS3 at the temperatures
In this study, it has a great importance to find the mentioned above (Tab. 5).
closest values of kinematic viscosity for both measurements The two Ubbelohde viscometers with type 2
to present the relation of Ubbelohde viscometer and (K = 0.1 mm2/s2) were used for the determination of
Stabinger viscometer and provide the traceability of the kinematic viscosity values of APN26 at the temperatures
Stabinger viscometer with our own primary viscosity of 20 °C, 25 °C, 40 °C, and with type 1C (K = 0.03 mm2/s2)
measurement system. at the temperatures of 60 °C, 80 °C (Tab. 6).
 G.S. Sariyerli et al.: Int. J. Metrol. Qual. Eng. 9, 7 (2018) 3

Table 4. APN415 Reference liquid (Method 2). Table 7. APN7.5 Reference liquid (Method 1).

Temperature (°C) Kinematic U0 (n) (%) Temperature (°C) Kinematic U0 (n) (%)
viscosity (mm2/s) viscosity (mm2/s)
20 1362.0 2 20 12.4011 0.28
25 978.6 2 25 10.5055 0.28
40 407.6 2 40 6.7832 0.25
60 157.8 2 60 4.2448 0.23
80 73.77 2 80 2.9151 0.22

Table 5. APS3 Reference liquid (Method 1). Table 8. APN415 Reference liquid (Method 1).

Temperature (°C) Kinematic U0 (n) (%) Temperature (°C) Kinematic U0 (n) (%)
viscosity (mm2/s) viscosity (mm2/s)
20 4.5398 0.23 20 1361.7885 0.38
25 4.0089 0.23 25 977.2290 0.38
40 2.8948 0.22 40 405.9990 0.34
60 2.0287 0.22 60 157.8822 0.34
80 1.5155 0.20 80 73.2143 0.32

(k = 2). The uncertainty value of temperature measuring

Table 6. APN26 Reference liquid (Method 1). device was taken as 0.02 K and the standard uncertainty was
calculated as 0.01 K. The uncertainty of the dynamic
Temperature (°C) Kinematic U0 (n) (%) viscosity is 1% (k = 1). The overall uncertainty of the
viscosity (mm2/s) measurements (taking into account the purity and handling
20 61.7185 0.32 of the samples) is estimated to be 2%. This equipment has
also a vibrating-tube that permits measurements of the
25 49.6582 0.32
densities with an uncertainty of 0.0005 g/cm3 (k = 1) [2].
40 27.8119 0.29
60 14.9192 0.29 2.2 Flow time measurement
80 9.0941 0.28
Flow-time measurements were performed by visual obser-
vation of the meniscus passing two ring marks at the
The two Ubbelohde viscometers with type 1C viscometer using a stopwatch with resolution of 0.001 s [3].
(K = 0.03 mm2/s2) were used for the determination of
kinematic viscosity values of APN7.5 at the temperatures 2.3 Temperature measurement
of 20 °C, 25 °C, 40 °C, and with type 1 (K = 0.01 mm2/s2) at
the temperatures of 60 °C, 80 °C (Tab. 7). The room temperature of viscosity laboratory of TUBI-
The two Ubbelohde viscometers with type 3C TAK UME is 20 °C ± 1 °C.
(K = 3 mm2/s2) were used for the determination of A thermostatically controlled bath of 70 L is used in the
kinematic viscosity values of APN415 at the temperatures measurements of viscometers with external cooling bath to
of 20 °C and 25 °C, with type 3 (K = 1 mm2/s2) at the provide the thermal equilibrium of measurements. Its
temperatures of 40 °C, with type 2B (K = 0.5 mm2/s2) at performance was tested in the range from 10 °C to 100 °C .
the temperatures of 60 °C and with type 2 (K = 0.1 mm2/s2) Temperature stability is 0.01 °C.
at the temperatures of 80 °C (Tab. 8). Temperature measurements are realized by using two
The internal diameter of each capillary has been SPRT standard platinum resistance thermometers with
manufactured in accordance with the German standard the sensitivity of 0.1 V/°C.
DIN 51562-1:1999; 283-mm length and 18 mL measuring
volume. UME Viscosity Laboratory maintains its own 3 Experimental setup
primary viscosity scale, based upon the internationally
agreed value of distilled water at 20 °C of 1.0034 mm2/s. First, the Ubbelohde viscometers were positioned in
Method 2, the Stabinger viscometer was used to the bath vertically and waited for thermal equilibrium.
determine the viscosity values of reference liquids. The For each reference liquids, the kinematic viscosity
apparatus was calibrated by the manufacturer, using several values were determined by the measurement method
density and viscosity standard reference fluids. The mentioned in the standard of DIN 51 562-1 and DIN 51
uncertainty of the temperature measurement is 0.02 K 562-4. The primary viscosity measurements were
4 G.S. Sariyerli et al.: Int. J. Metrol. Qual. Eng. 9, 7 (2018)

The kinematic viscosity values of reference liquids

(APS3, APN26, APN7.5 and APN 415) were measured
with a SVMTM3001 Stabinger viscometer (Anton-Paar) at
20 °C, 25 °C, 40 °C, 60 °C and 80 °C. This instrument offers
a 0.2–10 000 mPa s with a reproducibility 0.35% of the
measured value (Fig. 2).

4 Measurements and results

The viscosities were measured using either Ubbelohde
viscometers or an automated SVM 3001 Anton Paar
Stabinger viscometer.
The measurements were realized with 8 groups of
Ubbelohde type two glass capillary viscometers. The
uncertainty value of each kinematic viscosity was calcu-
lated according to DIN 510562-4 standard.
For the relative variance of the viscosity, a value
determined with two viscometers and using two ther-
mometers was [4,5,6]

1
ðS 0n Þ2 ¼ ðS 0K Þ2 þ ⋅½ðS 0Uhr Þ2 þ ðS 0#;V Þ2 þ ðS 0N Þ2
2
þ ðS 0t Þ2 ; ð1Þ

U 0n ¼ 2⋅S0n k ¼ 2:

The contributions of these quantities must have been

taken into account for the determination of the measure-
ment uncertainty. The respective relative empirical
standard deviations and the relative variance supplied
Fig. 1. Primary viscosity measurements system (Method 1). with the corresponding index were indicated as S0 and S0 2
respectively [4,5].
-Viscometer constant, S0 K2
If the relative uncertainty of the constant U0 K was
stated according to the calibration certificate for k = 2, the
variance was given by [4,5].

1
ðS 0K Þ2 ¼ ðU 0K Þ2 : ð2Þ
4
-Timing device, S0 Uhr2
It was assumed that two timing devices were used
corresponding to the usual practice to perform measure-
ments with both viscometers parallel in time in the same
thermostat bath [4,5].

1
Fig. 2. Stabinger viscometer (Method 2). ðS0Uhr Þ2 ¼ ðU Uhr Þ2 : ð3Þ
3
-Temperature measurement, S0 u,n2
realized by Ubbelohde viscometers seen in Figure 1. The contribution of the temperature measurement (Su)
Before all these measurements, the primary viscom- to the uncertainty of the viscosity determination included
eters had been characterized and their viscometer the calibration uncertainty of the temperature measuring
constants were determined. The calibration of the pairs device at the particular measuring temperature as well as
of viscometers was usually performed by simultaneous the temperature gradients in the thermostat bath, and was
direct comparison of two pairs of viscometers in the given by,
same temperature controlled bath. The different
liquids with known viscosities were used for the 1
ðS0#;n Þ2 ¼ ðS# ⋅U v Þ2 : ð4Þ
realization of procedure. 3
 G.S. Sariyerli et al.: Int. J. Metrol. Qual. Eng. 9, 7 (2018) 5

Fig. 3. Graphics of Kinematic viscosity value and temperature for reference liquids APS3, APN7.5 and APN26 in Method 1.

Fig. 4. Graphics of Kinematic viscosity value and temperature for reference liquids APS3, APN7.5 and APN26 in Method 2.
6 G.S. Sariyerli et al.: Int. J. Metrol. Qual. Eng. 9, 7 (2018)

Fig. 5. Graphics of Kinematic viscosity value and temperature for reference liquids APN415 in Method 1.

Fig. 6. Graphics of Kinematic viscosity value and temperature for reference liquid APN415 in Method 2.
 G.S. Sariyerli et al.: Int. J. Metrol. Qual. Eng. 9, 7 (2018) 7

Table 9. Relative variance of flow time depending on n rotor. At equilibrium, the two torques are equal and
and et. the viscosity can be traced back to a single speed
2 measurement. The SVM 3001 uses Peltier elements for
ðet ⋅F Þ
n et F S 02
t ¼ n fast and efficient thermal stability. The temperature
uncertainty is ±0.02 K (k = 2) from 288.15 to 378.15 K.
3 1 · 10
3 0.591 11.6 · 10
8
The precision of the dynamic viscosity measurements
3 2 · 10
3 0.591 46.7 · 10
8 is ±0.5%. The overall uncertainty of the measurements
4 1 · 10
3 0.486 5.9 · 10
8 (taking into account the purity and handling of the
4 2 · 10
3 0.486 23.6 · 10
8 samples) is estimated to be 2% [2,7–9].
5 1 · 10
3 0.430 3.7 · 10
8
5 2 · 10
3 0.430 14.9 · 10
8
5 Conclusion
-Inclination error, S 0 N2
The viscometer was inserted into a holder equipped In this paper, the viscosity measurements were realized
with a flat top plate which served for the defined fixation of both glass capillary viscometers and Stabinger viscometer
the viscometer in the thermostatic bath. The deviation of by using certified reference liquids. In first method, the
the capillary axis from the vertical to the top plate of the Ubbelohde viscometers were used to find the flow time of
holder must not have exceeded 2 [4,5]. liquids and then the kinematic viscosity was calculated. In
second method, the Stabinger viscometers measured the
dynamic viscosity and the kinematic viscosity was found

2
 using density and dynamic viscosity [6].
1 C
C0 1 cosu 2
ðS0N Þ2 ¼ ¼ 1
; ð5Þ The measurement results of kinematic viscosity values
3 C 3 cosu0 for both methods were so close as it could be seen in
Figures 3–6.
cosu The evaluation of expanded uncertainty of Method 1 is
C 0 ¼ C⋅ u ¼ 2° ; u0 ¼ 0: mainly affected by the individual contributions of the
cosu0
viscometer constant and flow time.
-Flow time, S0 t2 The main idea of this study that using Stabinger
This contribution comprised deviations due to visual or viscometer to measure the viscosity value of liquids is
automatic recording of the meniscus passages as well as easy and accurate way due to the high uncertainty value.
changes in temperature as to time. Providing the traceability by Ubbelohde viscometers, the
For each reference liquids, the mean value (t) of the flow results can be used to give certificate for the reference
time (ti) was calculated as below. As well as the relative liquids and calibration of viscometers. This paper showed
difference between the longest and shortest flow time was the procedure of the determination of viscosity value of
determined [4,5]. each liquids by using Ubbelohde viscometers and
Stabinger viscometer at different temperatures. The
values are very similar that the evaluation can be realized
1X tmax
tmin in uncertainty budget of the Stabinger viscometer.
t¼ ti et ¼ : ð6Þ According to this, the evaluation can be given to
n t
decrease the uncertainty values of Method 2 in future
n as the number of flow times of measurement series, et as study.
the relative range of flow in measurement series of a
viscometer and F as multiplication factor were given in
Table 9 [4,5].
Measurements of viscosity were also performed in the References
temperature range of 20 °C–80 °C at atmospheric pressure
using an automated SVM 3001 Anton Paar Stabinger 1. J.G. Webster, Mechanical Variables Measurement Solid,
viscometer. Fluid and Thermal, CRC Press LLC, USA, 2000
These viscosity measurements are based on a tube 2. X. Parades, O. Fandino, A.S. Pensado, M.J.P. Comunas, J.
Fernandez, Experimental density and viscosity measure-
filled with the sample in which floats a hollow
ments of di(2ethylhexyl)sebacate at high pressure, J. Chem.
measuring rotor. Due to its low density, the rotor is
Thermodyn. 44, 38–43 (2012)
centered in the heavier liquid by buoyancy forces. The 3. S. Lorefice, F. Saba, The Italian primary kinematic
rotor is forced to rotate by shear stresses in the liquid viscosity standard: the viscosity scale, Measurement 112,
and is guided axially by a built-in permanent magnet, 1–8 (2017)
which interacts with a soft iron ring. The rotating 4. DIN 51562-1, Determination of the Kinematic Viscosity Using
magnetic field delivers the speed signal and induces the Standard Design Ubbelohde Viscometer (German Insti-
eddy currents in the surrounding copper casing. These tute for Standardization DIN, Berlin, Germany, 1999)
eddy currents are proportional to the speed of the rotor 5. DIN 51562-4, Measurement of Kinematic Viscosity by means
and exert a retarding torque on the rotor. Two of the Ubbelohde Viscometer (German Institute for Standard-
different torques influence the speed of the measuring ization DIN, Berlin, Germany, 1999)
8 G.S. Sariyerli et al.: Int. J. Metrol. Qual. Eng. 9, 7 (2018)

6. H. Bauer, G. Klingenberg, EUROMET.M.V-K3 Key Inter- 8. S. Geacai, O. Iulian, I. Nita, Measurement, correlation and
comparison of Liquid Viscosity Measurements (PTB-ThEx- prediction of biodiesel blends viscosity, Fuel 143, 268–274 (2015)
22, Bericht, 2001), p. 75 9. J.L. Camas-Anzuetoa, J. Gómez-Péreza, R. Meza-Gordilloa,
7. D.H. Lee, J.M. Jung, S.Y. Kim, K.T. Kim, Y.I. Cho, G. Anzueto-Sánchezb, M. Pérez-Patricioa, F.R. López-
Comparison tests for plasma viscosity measurements, Int. Estradaa, M. Abud-Archilaa, C. Ríos-Rojasa, Measurement
Commun. Heat Mass Transf. 39, 1474–1477 (2012) of the viscosity of biodiesel by using an optical viscometer,
Flow Meas. Instrum. 54, 82–87 (2017)

Cite this article as: G.S. Sariyerli, O. Sakarya, U.Y. Akcadag, Comparison tests for the determination of the viscosity values of
reference liquids by capillary viscometers and stabinger viscometer SVM 3001, Int. J. Metrol. Qual. Eng. 9, 7 (2018)