Deliverable D1.1
Title: Testing methodology manual
Editor
Participants
Issue date
Dr.N. Stryzhakova
Dr.Y. Maletin, Dr.S.Zelinskyi
13.03.2012
Dissemination
Public
Restricted to other programme participants (including the Commission
Services)
RE Restricted to a group specified by the consortium (including the Commission
Services)
CO Confidential, only for members of the consortium (including the Commission
Services)
PU
PP

FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
1
Table of Contents
1.
Abstract ........................................................................................................................................... 3
2.
Introduction .................................................................................................................................... 4
3.
Activities, results ............................................................................................................................. 5
3.1
3.1.1.
Capacitance ..................................................................................................................... 5
3.1.2.
Internal resistance .......................................................................................................... 8
3.1.3.
Constant current method ............................................................................................... 9
3.1.4.
Impedance .................................................................................................................... 10
3.2
4.
5.
Test procedures for supercapacitors ...................................................................................... 5
Specific energy and specific power ....................................................................................... 11
3.2.1.
Constant Power Tests [5] .............................................................................................. 11
3.2.2.
Maximum energy stored ............................................................................................... 12
3.2.3.
Maximum Power ........................................................................................................... 12
3.3
Self-Discharge Test [1, 2, 5]................................................................................................... 12
3.4
Cycle-Life Tests [5] ................................................................................................................ 13
3.5
Endurance test [1, 2, 5] ......................................................................................................... 14
Test procedures for hybrid supercapacitor (HS) cell and stack .................................................... 16
4.1
Constant current tests .......................................................................................................... 16
4.2
Constant power tests ............................................................................................................ 18
4.3
Self-Discharge Test ................................................................................................................ 20
4.4
Cycle-Life Tests ...................................................................................................................... 20
4.5
Endurance test ...................................................................................................................... 21
Examples of experimental results ................................................................................................. 22
5.1
Constant current cycling ....................................................................................................... 22
5.2
Constant power cycling ......................................................................................................... 26
6.
Conclusions ................................................................................................................................... 29
7.
References .................................................................................................................................... 30
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
2
1. ABSTRACT
Analysis of standard testing procedures to measure parameters of double-layer capacitors
(supercapacitors) was carried out. This analysis included procedures proposed by the International
Standards IEC 62391-2 and IEC 62576, USABC, UC Davis as well as procedures used by Yunasko.
Testing methodology was developed for testing hybrid supercapacitor cells and stacks.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
3
2. INTRODUCTION
The main Scientific Objective of Energy Caps is the development of a sustainable and safe
hybrid supercapacitor with high specific energy of 100Wh/kg and maintained high specific power of
10kW/kg and cyclability of > 105 cycles for electrical storage application in electric cars, plug-in
hybrids and the smart grid. This objective will be realised by development of optimized
electrochemical systems including graphite-based electrode compositions, activated carbon
electrode compositions, new separators and electrolyte.
Material properties are studied using cells that are lab prototypes with one pair of
electrodes (positive and negative). The main characteristics of cell are: rated voltage, capacitance,
internal resistance, specific energy and specific power.
After the selection of optimized material compositions testing is done on larger surface cells
or stacks that are devices with set of positive and negative electrodes connected in parallel within
the same case (internal connection). That way allows increasing capacitance and decreasing
resistance. Stacks may be connected in serial and/or parallel by external combination to achieve the
desired performances. The main characteristics of stack are: rated voltage, capacitance, internal
resistance, specific energy, specific power, specific energy vs. specific power (Ragone plot),
temperature dependence of resistance and capacitance (-30÷70 ºC), cycle life, self discharge,
calendar life (hours) at fixed voltage and high temperature (40-60 ºC).
In the frame of WP1 the analysis of standard testing procedures to measure parameters of
double-layer capacitors (supercapacitors) was carried out. This analysis included procedures
proposed by the International Standards IEC 62391-2 and IEC 62576, USABC, UC Davis as well as
procedures used by Yunasko. Testing methodology was developed for testing hybrid supercapacitor
cells and stacks and test manual is proposed as results of this work.
List of test procedures to be used for determining the main characteristics of
supercapacitors
•
Constant current charge/discharge
Capacitance and resistance
Cycle life
•
Pulse tests to determine resistance
•
Constant power charge/discharge
Determine the Ragone Plot for power densities between
100 and at least 1000 W/kg
for the voltage between Ur
and ½ Ur.
Test at increasing W/kg until discharge time is less than 5 sec. The charging is often carried
out under constant current conditions with a charging time of at least 30 sec.
•
Voltage maintenance
Self discharge test
•
Continuous application of rated voltage at high temperature
Endurance test (calendar life).
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
4
3. ACTIVITIES, RESULTS
3.1 Test procedures for supercapacitors
3.1.1. Capacitance
The capacitance shall be measured by using the constant current charging and discharging
methods. Figure 1 shows the basic circuit to be used for the measurement [2].
Fig.1 – Basic circuit for measuring capacitance and resistance:
ICC – constant current
UCC – constant voltage
- d.c. ammeter
- d.c. voltage recorder
S – changeover switch
Cx – capacitor under test
- constant current discharger
a) constant current charging
b) constant voltage charging
The test equipment shall be capable of constant current charging, constant voltage charging,
constant current discharging, and continuous measurement of current and voltage through the
capacitor terminals. The test equipment shall be able to set and measure the current and voltage
values with the accuracy of ± 1 % or better.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
5
The power supply shall provide the constant charge current for the capacitor charge with 95 %
efficiency, set the duration of constant voltage charge, and provide a discharge current
corresponding to the specified discharge efficiency. The d.c. voltage recorder shall be capable of
conducting measurements and recording with a 5 mV resolution and sampling interval from 10 to
100 ms.
In general, before starting measurements, a detailed test plan should be developed, which
includes, in particular, the selection of test conditions appropriate for evaluating the device. For
instance, for dc testing this includes selection of current values (in A) at which the device will be
charged and discharged, as well as the initial and final voltages for the tests. Hereinafter are several
approaches for this purpose.
a) USABC test procedure
The USABC test manual [3] specifies a series of constant current tests to characterize the
supercapacitors, but the tests are specified in terms more appropriate for batteries than for
supercapacitors [4]. By that is meant that the discharge rates are given as nC based on an effective
Ah rating, where Ah=C(Vmax– Vmin)/3600. The rated capacitance C is measured at the 5C rate, which is
a very low rate for a supercapacitor (NOTE: 5C rate corresponds to 12 minute discharge). The current
values to be used for this test are specified fractions of IMAX as shown in the “Normal Test Currents”
column of Table 1. The Rated Maximum Current (IMAX) is normally specified by the device
manufacturer or developer. Every effort should be made to get a realistic value for this parameter
from the supplier. If this is not possible, a default value of IMAX can be established as the smallest of
the following: (a) the current required to cause an immediate (i.e., <0.1 s) 20 % voltage drop in a
fully charged device at 30 ºC, or (b) the current required to discharge the device from UMAX to UMIN in
2 s. Sometimes test equipment limitations can decrease this full range of currents. In this case, the
maximum available current has to be chosen as the maximum test current, and the sequence of test
current values is then as shown in Table 1in the “Test Equipment Limited” column.
Table 1. Constant-Current Discharge and Charge Test Values [3]
Normal Test Currents
(Discharge & Charge)
Minimum Test Current
Other Test Currents
Maximum Test Current
5C
0.1 IMAX
0.25 IMAX
0.5 IMAX
0.75 IMAX
IMAX
Test Currents
Test Equipment Limited
to ITEST < IMAX
(Discharge & Charge)
5C
0.1 ITEST
0.25 ITEST
0.5 ITEST
0.75 ITEST
ITEST
b) IEC procedure
The IEC 62576 standard [2] describes a test procedure for determining the capacitance C and
resistance R of supercapacitors. This procedure specifies a single test intended to determine the
performance of the capacitor at a single current – so that the efficiency in charge and discharge to
be 95%. It is assumed that the capacitors to be tested behave as ideal double-layer capacitors with
constant C and R values. The resultant relationships for the test current values are the following:
I ch 
Ur
38R
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
6
I dch 
Ur
40 R
C value (in F) is calculated from the discharge part of a cycling curve by the equation:
C
I dch  t
U
where:
Idch is the discharge current, A
∆t is the discharge time, sec
∆U is the change in potential, V, that is chosen as follows: U  0.9U r  0.7U r
If the supercapacitor behavior is not ideal, and the U vs t plot is deviated from the straight
line in the voltage range chosen, then capacitance value should be calculated by “energy
conversation capacitance method”
C
2E
,
(0.9U r )  (0.7U r ) 2
2
where E is the energy (in J) obtained from a capacitor during discharge in this voltage range. It is
calculated by integrating the product of voltage by current for each time increment during the
discharge.
c) UC Davis test procedure
Dr. A.Burke (UC Davis, CA) uses the nominal charge/discharge current In. The nominal
charge/discharge current In can be determined from the rated capacitance and voltage of the device
with the use of the following relationship [4-6]:
where Pn is the nominal power density (W/kg) found in tests with the use of 200 or 400 W/kg;
m is the device weight (in kg);
Ur is the maximum d.c. voltage (in V) that may be applied continuously for a certain time under the
upper category temperature to a capacitor so that a capacitor can exhibit specified demand
characteristics.
Selection of the power levels for tests would depend on the mass (m) of the device to be
evaluated.
Before measurements, the capacitors shall be fully discharged and then incubated for 2 h
under the reference temperature, set at 25 °C ± 2 °C, as specified in IEC 60068-1, clause 5.2, or that
specified by the related standards.
Tests will then be performed at currents higher and lower than In (e.g., 0.25, 0.5, 1.0, 2.0, 4.0,
8.0In). Three consecutive charge/discharge cycles at a specified current value constitutes a single
test. The reported data should be the average of the second and third cycles. The voltage is held for
20 s at the end of the charge and discharge portions of the test cycle to establish a constant voltage
level for all tests.
The capacitance C shall be calculated using the equation below, which is based on the voltagetime characteristics between capacitor terminals
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
7
where C is the device capacitance (F)
Itest is the test current (A)
ttest is the discharge time (s)
Ur is the rated voltage (V)
Umin is the minimal discharge voltage (V)
d) YUNASKO test procedure
In our lab we normally use the testing procedure that includes constant current, constant
voltage charge/discharge, a set of the current values being in range from 0.2Itest to Itest.. Capacitance
value C (in F) is calculated from the discharge part of a cycling curve by the equation:
C
I test  t
U
where:
Itest is the discharge current, in A
∆t is the discharge time, in s
∆U is the voltage change, in V, that is chosen as follows:
U  0.9(U r  U drop)  0.7(U r  U drop)
From the experimental plot of capacitance vs current: С=f(I), the maximum capacitance
value С0 (as a value at the low current tending to zero) and -dC/dI value (the slope) can be found
with the use of the least square technique, and these values can then be used to evaluate the
maximum energy stored and power capability. It should be noted that the parameter -dC/dI is an
important one, because it can characterize the system behavior at high power load.
3.1.2. Internal resistance
Determination of the resistance of the capacitor is complicated by the fact that the voltage
decreases due to both the internal resistance and loss of charge. In addition, due to the porous
character of the electrode, the resistance of the capacitor varies with time until the charge
distribution within the electrodes is fully established [4].
The internal resistance can be represented as an Equivalent Series Resistance (ESR) or as
Equivalent Distributed Resistance (EDR). ESR is the so-called pulse resistance corresponding to all the
resistive components within the supercapacitor. EDR value includes ESR and also includes an
additional contribution of the charge redistribution process in the electrode pore matrix that takes
place at any voltage jump or drop due to inhomogeneous electrode structure, the process adding
significantly to Joule heating or ohmic heating the supercapacitor (evaluated as I2Rt).
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
8
3.1.3. Constant current method
ESR value, when charging/discharging the supercapacitor with the constant current I, or RDС
(in Оhm) can be calculated from the value of voltage jump/drop (Ujump/drop=U4 – see Fig. 22), when
the current direction changes from discharge to charge or vice versa in accordance with the
following equation:
RDC 
U jump / drop
I
where :
∆I is the difference between current values before (I1) and after (I2) the current direction changes ∆I
= |I1 - I2|; the time interval between two current measurements (sampling rate) does not exceed 10
ms.
For many supercapacitor applications, this is the steady-state resistance, which is the most
relevant value for the calculation of power capability/electrical losses/heating, but not the R0 (ESR)
value, which is typically less than the steady-state one. It is important to define what resistance
value is being reported [4].
EDR is determined according to IEC 62576 [2] or IEC 62391-1 [1]. Test procedure is as
follows:
- The voltage-time plot is measured under galvanostatic charge-discharge;
- The straight-line approximation to the voltage drop characteristics from the start voltage
(0.9(Ur-Udrop)) to end voltage (0.7(Ur-Udrop)) is applied with the use of the least squares method;
- The interception of the straight line with the vertical line at the moment when the discharge
is switched on gives the voltage drop, U3 (see Fig. 2); from which the EDR value can be calculated
using the equation:
EDR 
U 3
I dch
In this procedure Idch is the current at 95% efficiency (see the chapter 1.1.2 above).
Fig. 2. Measurement profile for ESR (from U4) or EDR (from U3) evaluation.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
9
As our practice show, the reliable values of ESR can be obtained if ΔU is determined from the
readings at the end of the 20 s period of holding the voltage prior to charge or discharge, and within
10 ms voltage jump or drop after the initiation of charge or discharge, respectively. The ESR should
be calculated for the beginning of charge or discharge for each of the constant-current test cycles. It
should be independent of current. On the other hand, EDR values depend on current values – see
Fig.3.
Fig.3 – Plots of ESR and EDR vs. normalized current values.
Based on the test results it was shown [6] that the current interruption approach is
recommended for determining the resistance of batteries and supercapacitors. The resistances of
both types of devices are time dependent due to physical and electrochemical processes occurring in
the electrodes. For the battery, the use of the current interruption method gives Rs = U/I, where
U should be taken at 10 ms if the Ohmic resistance is needed and at 2 s if the resistance including
Ohmic and limited electrochemical processes are appropriate for the application of interest. For the
supercapacitor, it is recommended that the voltage time be somewhat less than 10 ms, possibly 1
ms when the test equipment is capable, following the current interruption I –> 0. It is difficult to
utilize the 0 –> I pulse method to determine the resistance without carefully accounting for the
capacitance response of the device. For that reason, the current interruption method is particularly
recommended for supercapacitors.
3.1.4. Impedance
Impedance spectra can be used to determine the equivalent series resistance [5]. The ac
resistance at 1 kHz can be a good parallel reference for dc resistance measurement at 1 ms time
frame [6].
Impedance spectra can also identify the self-resonance frequency of a device. Importantly,
impedance data can be used to create a two-terminal equivalent circuit which models the electrical
response of the device in detail.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
10
The device should be conditioned at the specified test temperature until thermal equilibrium
is reached. Initialize the device voltage at the specified voltage for a minimum of 1 hr before each
test. Longer times may be necessary to stabilize the voltage on some devices. Test voltages are Ur,
(Ur+Umin)/2, and Umin. Measure the device impedance at a minimum of 5 frequencies per decade,
over the range of 0.001 Hz to above the frequency of at least 1000 Hz. The amplitude of the applied
sinusoidal signal must be set so that the average value applied to each cell is less than or equal to
0.02 V. The real and imaginary parts of the impedance at each measurement frequency shall be
recorded
Data should be presented on a log-log plot of the magnitude of the real and the imaginary
part of the impedance versus frequency at each of the voltages. Three frequencies should be
identified:
1. The frequency where the real and the imaginary parts of the impedance are equal.
2. The frequency where deviations from ideal-capacitor behavior take place.
3. The self-resonance frequency.
3.2 Specific energy and specific power
3.2.1. Constant Power Tests [5]
This procedure specifies a method to determine the discharge characteristics of the device at
different discharge power levels. The range of discharge rates (powers) is selected so that they
correspond to power densities (Pm) between 200 and 1200 W/kg (or higher for advanced devices).
The discharge is performed between Ur and Umin. The device should be charged at constant-power of
200W/kg and held at Ur for 20 s before the constant power discharge is initiated. Three consecutive
charge/discharge cycles at each specified constant power constitutes a single test.
Based on the rated voltage Ur and the mass m of the device, discharge power values are
calculated for specific power densities of 200, 500, 1000, 2000 W/kg. Tests are performed at each of
these discharge power values. The discharge times for these tests should be in the same range as the
constant current tests. For each constant power test, the energy E during charge and discharge will
be calculated by summing U∙I∙Δt during the charge/discharge portions of the cycle. The usable
energy densities Em (Wh/kg) and Ev (Wh/l) are calculated from
∑
∑
and the efficiency η is calculated from
∑
∑
A plot illustrating Em (or Ev) vs Pm (or Pv) is known as Ragone plot.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
11
3.2.2. Maximum energy stored
This is the energy (in W.h), which a supercapacitor can store at charging up to the rated
voltage. It can be calculated using the following equation:
Emax
CU r2

(Wh / kg)
2  3600m
Available energy at discharge from the rated voltage Ur to Ur/2 is equal:
Eavail
3CU r2

(Wh / kg)
8  3600m
3.2.3. Maximum Power
Maximum power is the power that can be delivered to the load from a charged
supercapacitor. According to IEC 62576 it is calculated by using the internal resistance value (EDR)
and the following equation:
Pmax
0.25U r2

(W / kg)
Rm
where:
Pmax is the maximum power of supercapacitor (W);
Ur is the rated voltage (V);
R is the equivalent distributed resistance (Ohm);
m is the mass of device (kg).
This calculation method is known as “matched impedance power density method”.
A different equation can be used if one needs to calculate the power output at certain
efficiency and at discharging the supercapacitor from the rated voltage Ur to Ur/2:
P 
9(1   )U r2
16 Rm
where:
Pn is the power output at the efficiency of η (W);
Ur is the rated voltage (V);
R is the equivalent distributed resistance (Ohm).
It should be noted that in YUNASKO calculations the η value is normally 0.95 if a different
value is not specified.
3.3 Self-Discharge Test [1, 2, 5]
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
12
The self-discharge test measures the time dependence of the self-dissipation of the
capacitor, i.e., the rate of those internal processes that cause the capacitor to discharge when not
connected to a load. The goal of the test is twofold: (1) to measure (directly) the energy loss over the
test interval; and (2) to measure the decrease of the capacitor’s voltage during the test, which allows
the associated energy loss to be calculated as a function of stand time.
The self-discharge measurement profile is shown in Fig. 4 [1].
Fig. 4. Self-discharge test diagram.
Measured parameter is the voltage maintenance rate (B), which is calculated by the
following equation:
B
U end
 100%
Ur
where:
B is the voltage maintenance rate (%);
Uend is the voltage between open capacitor terminals after 72 h;
Ur is the rated voltage.
3.4 Cycle-Life Tests [5]
Stable performance during more than 100,000 charge/discharge cycles is desired. This
procedure measures device performance stability during cycle tests. The devices are characterized
initially and then periodically throughout the cycle test. Individual device failures and the number of
cycles completed before the failure should be recorded. Constant-current charging and discharging
are used.
Typical procedure: Condition the capacitor at 25 ± 3°C until thermal equilibrium is reached.
Charge the device at a current
In 
Ur
40 R
from Ur/2 to Ur so that the voltage reaches Ur in 30 (± 1)
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
13
s. Maintain voltage Ur on the device for 15 ± 0.50 s. Then discharge the capacitor to Umin at current I.
Hold at Umin for 50 ± 0.50 s. This defines a cycle. Repeat this cycle throughout the testing, adjusting I
as needed in order to maintain the initial charge/discharge times. Devices shall be characterized
using the procedures listed below initially, and after 1000 ± 25; 4000 ± 100; 10,000 ± 250; 40,000 ±
1000; 100,000 ± 2500; 150,000 ± 2500; and 200,000 ± 2500. Device failure and the number of cycles
before the failure shall be recorded. Failed devices should be removed from tests.
Characterization tests to be performed at each measurement cycle include:
1. Constant-Current Charge/Discharge (In)
2. ESR (from constant-current test data)
3. Constant Power Discharge (200 W/kg, 1000 W/kg)
1.6 Temperature Performance [5]
Temperature influences the energy that can be stored in a device as well as the power it can
deliver. Charge/discharge cycle efficiency is also dependent on temperature. Device performance
measurements are carried out at three temperature values; in this procedure:
Step 1 - Condition the device at 25 + 3°C and perform the tests listed below. Data from
previous measurements at this test temperature may be used provided that they were acquired
within the previous thirty (30) days.
Tests
1. Constant-Current Charge/Discharge (In)
2. ESR (from constant-current test data)
3. Constant Power Discharge (200 W/kg, 1000 W/kg)
Step 2 - Condition the capacitor at 60 ± 3°C until thermal equilibrium is achieved. Perform the
above mentioned tests at this temperature in the order listed.
Step 3 - Condition the capacitor at -30 ± 3°C until thermal equilibrium is achieved. Perform the
above mentioned tests at this temperature in the order listed.
Step 4 - Condition the capacitor at 25 ± 3°C and repeat the tests listed above. This test data
will provide information about the stability of the capacitor under thermal cycling conditions.
Step 5 - Perform a visual inspection of the capacitors to identify any damage or electrolyte
leakage caused by the thermal cycle.
3.5 Endurance test [1, 2, 5]
This procedure characterizes device life properties and performance using an accelerated
aging condition.
Device properties and performance are measured initially and then periodically throughout
the aging period. Individual device failures and the times to failure are recorded. Voltage equal to
the rated voltage, Ur, is applied and maintained during specified characterization tests.
Age the capacitors in a suitable oven or environmental chamber maintained at 60 ± 3°C with
an applied voltage equal to Ur. The voltage source must be capable of supplying a current of at least
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
14
ten times more than the steady state current draw of the capacitor at 60°C. This will ensure that the
device failure is not limited by the tester or cell short circuiting takes place.
Characterization tests of the devices should be performed at the start of the test sequence
and after 250 ± 10, 500 ± 25, 1000 ± 50, and 2000 ± 100 hours. Measurements are made at 25 ± 3°C.
Device failures and the time to failure should be recorded. Failed devices should be removed from
tests. The characterization tests to be performed at each measurement time include:
1. Constant-Current Charge/Discharge (In)
2. ESR (from constant-current test data)
3. Constant Power Discharge (200 W/kg, 1000 W/kg)
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
15
4. TEST PROCEDURES FOR HYBRID SUPERCAPACITOR (HS) CELL
AND STACK
4.1 Constant current tests
The capacitance and resistance values can be measured by cycling a HS device with a constant
current from the rated (upper) voltage Umax = Ur to some lower value, typically to Umin = Ur/2 (Ur –
rated voltage).
The test equipment shall be capable of constant current charging, constant current
discharging, and continuous measuring the current and voltage values as the HS terminals. The test
equipment shall be able to set and measure the current and voltage values with the accuracy of ± 1
% or better. The d.c. voltage recorder shall be capable of conducting measurements and recording
with a 1 mV resolution and sampling interval from 10 to 100 ms.
Current value selection. The rated capacitance Cr is measured at the 5C rate, which
corresponds to 12 minute discharge (I5C). This capacitance value will be used as reference one.
The constant-current charging and discharging characteristics at increasing currents are
determined by executing a sequence of discharge/charge cycles. A set of the current values is in the
range 2I5C; 4I5C; 8I5C ... Imax and Imax corresponds to 5s discharge.
Test sequence (see Fig.5)
1. Fully charge the device to Umax (U1, t1).
2. Place the device in an open circuit condition for time equal to ~5RC (t2-t3).
3. Discharge the device to Umin at the lowest test current (U3, t3 – U5, t5).
4. Place the device in an open circuit condition for time equal to ~5RC (t6 –t7).
5. Charge the device to Umax at the lowest test current (U7, t7 – U9, t9).
6. Repeat Steps 2 through 5 for 4 more iterations (5 cycles total).
Repeat the entire sequence of Steps 1 through 6 for each test current. Each of the 6 test
current sequences should begin at a fully charged, stable state within the normal temperature limits
(i.e., 25±2 ºC for an ambient temperature test).
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
16
Fig.5 – Charge/discharge cycles for the Ioxus hybrid device RHE2R3108SR (voltage in black,
absolute value of current in blue).
Data processing and parameter calculation
Capacitance (C). HS charge/discharge curves are not expected to be linear; therefore
capacitance should be calculated by the energy conversion capacitance method:
C
2E
U4 U5
2
2
where
C is the capacitance (in F);
E is the energy (in J) delivered at discharging from the voltage U4 to the voltage U5. This value is
calculated as follows:
t5
E  I  U (t )dt
t4
After testing a set of capacitance values as a function of current values can be obtained. From
the plot of capacitance vs current: С=f(I), the maximum capacitance value С0 (as the value at current
tending to zero) and -dC/dI value (the slope) can be found with the use of the least square
technique, and these values can then be used to evaluate the maximum energy stored and power
capability. It should be noted that the -dC/dI parameter is an important one, since it can characterize
the system behavior at high power load.
Testing report. Rated capacitance Cr (at 5C) and the plot of capacitance vs. current.
Internal resistance. The internal resistance can be presented as Equivalent Series Resistance
(ESR) or as Equivalent Distributed Resistance (EDR). ESR is the so-called pulse resistance
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
17
corresponding to all the resistive components within the HS. EDR value includes ESR and also
includes an additional contribution of the charge redistribution process in the electrode pore matrix
that takes place at any voltage jump or drop due to inhomogeneous electrode structure.
ESR value is calculated from the value of voltage drop (see Fig.1):
ESR 
U3 U 4
I dch
where Idch is the value of discharge current.
EDR is determined from the straight-line approximation in the voltage range 0.9U4 – 0.7U4 as
shown in Fig.6:
EDR 
U
I dch
Fig.6 – EDR definition
Testing report. Plots of ESR and EDR vs. currents.
4.2 Constant power tests
This procedure specifies a method to determine the discharge characteristics of the device
at different discharge power levels. The range of discharge rates (powers) is selected so that they
correspond to power densities (Pm) between 200 and 2000 W/kg (or higher for advanced devices).
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
18
The discharge is performed between Ur and Umin. The device should be charged at constant-power of
200W/kg and held at Ur for the time interval ~5RC s followed by the constant power discharge is
initiated. A single test includes three consecutive charge/discharge cycles at each specified constant
power value.
Based on the rated voltage Ur and the mass m of the device, discharge power values are
calculated for specific power densities of 200, 500, 1000, 2000 W/kg. Tests are performed at each of
these discharge power values. The discharge times for these tests should be in the same range as the
constant current tests. For each constant power test, the energy E during charge and discharge will
be calculated by summing U∙I∙Δt during the charge/discharge portions of the cycle. The usable
energy densities Em (Wh/kg) and Ev (Wh/l) are calculated from
∑
∑
and the efficiency η is calculated from
∑
∑
Maximum energy stored is the energy (in W.h), which a HS device can store at charging up to
the rated voltage. It can be calculated using the following equation:
Emax
CrU r2

(Wh / kg)
2  3600m
Available energy at discharging from the rated voltage Ur to Ur/2 can be found as:
Eavail
3CU r2

(Wh / kg)
8  3600m
Maximum power is the power that can be delivered to the load from a fully charged HS
provided that the load resistance is equal to the internal resistance of HS (NOTE: this corresponds to
efficiency 0.5). According to IEC 62576 it is calculated by using the internal resistance value (EDR)
and the following equation:
Pmax
0.25U r2

(W / kg)
Rm
where:
Pmax is the maximum power of HS (W);
Ur is the rated voltage (V);
R is the equivalent distributed resistance EDR (Ohm);
m is the mass of device (kg).
This calculation method is known as “matched impedance power density method”.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
19
Testing report. A plot of Em (or Ev) vs Pm (or Pv) (known as Ragone plot) as well as maximum
energy and maximum power values.
4.3 Self-Discharge Test
Test sequence (see Fig.7)
1. Fully charge the device to Ur.
2. Keep the device at Ur for 2 hours.
3. Leave the device at open circuit conditions and measure the voltage over 72 hours.
Fig. 7. Self-discharge test diagram
Testing report. A plot of U vs. time as well as the voltage maintenance rate (B), which is
calculated by the following equation:
B
U end
 100%
Ur
where:
B is the voltage maintenance rate (%);
Uend is the voltage between open capacitor terminals after 72 h;
Ur is the rated voltage.
4.4 Cycle-Life Tests
The devices are characterized initially and then periodically throughout the cycle test.
Individual device failures and the number of cycles completed before the failure should be recorded.
Constant-current charging and discharging are used. Constant current should provide about 3 min of
discharging (20C current).
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
20
Test sequence
1. Charge the device to Ur.
2. Place the device in an open circuit condition for time about 3RC s.
3. Discharge the device to Umin=Ur/2.
4. Place the device in an open circuit condition for time about 3RC s.
5. Repeat Steps 1 through 4.
Devices should be characterized initially, and after 500, 1000, 2000, 5000, and then after
each 1000 cycles if possible. Device failure and the number of cycles before the failure shall be
recorded. Failed devices should be removed from tests. The device temperature should be under
control during cycling and cycling must be suspended when the temperature will exceed upper
temperature limit for the device (for example, this temperature limit is about 60 C for the
acetonitrile based device).
4.5 Endurance test
Device properties and performance are measured initially and then periodically throughout
the aging period. Individual device failures and time to failure are recorded. The rated voltage, Ur, is
applied and maintained during the specified characterization tests.
The capacitors are aged in a suitable oven or environmental chamber maintained at 60 ± 3°C
with an applied voltage equal to Ur. The power supply must be capable of supplying a current of at
least ten times more than the floating current of the capacitor at 60°C.
Characterization tests of the devices should be performed beforehand and then repeated
after 100, 250, 500, and 1000 hours. Measurements are made at 25 ± 3°C. Device failures and the
time to failure should be recorded. Failed devices should be removed from tests.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
21
5. EXAMPLES OF EXPERIMENTAL RESULTS
5.1 Constant current cycling
Testing methodology and procedures described above were used for testing supercapacitor
and hybrid stacks as testing samples. For this work we have chosen three SC stacks of different rated
parameters, design features and size, namely YUNASKO 320F, YUNASKO 1200F and Maxwell
Technologies 1500F devices as well as Ioxus hybrid cell of 1000 F and Yunasko H3-6000F hybrid
device, which is based on mechanical mixtures of the activated carbon with Li-ion battery
components.
Tests were carried out using home-made measuring stand (Fig8). Supercapacitors were
cycled in galvanostatic regime (constant current or constant current-constant voltage cycling) at 20,
40, 60, 80 and 100 A steps. Fig9-11 illustrate the typical charge-discharge curves for these samples.
The key parameters (capacitance and resistance) were obtained from these data. These values are
listed in Erreur ! Source du renvoi introuvable.2 below.
Fig. 8. Picture of YUNASKO measuring stand with supercapacitor stacks under testing.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
22
Fig. 9. Constant current cycling curves for Yunasko 330 F supercapacitor.
Fig. 10. Constant current cycling curves for Ioxus hybrid 1000 F supercapacitor.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
23
Fig. 11. Constant current cycling curves for Yunasko H3-6000 F hybrid device.
Based on the obtained graphs, the main rating parameters can also be determined for hybrid
devices, though in a bit different way:
-
Resistance is determined in the same way as for ordinary EDLCs. It is measured on the edge
between chargin-dischargin the device.
-
Capacity (versus “capacitance” for EDLCs) is measured using the following equation:
,
where
∫
-
Energy is determined by the equation:
,
(Q has the value of A∙h)
-
Power is calculated by the equation:
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
24
Table 2. Capacitance (C) and resistance (ESR, EDR) obtained from charge-discharge curves when
cycling with different current values.
Stack name
I, A
ESR, mOhm
EDR, mOhm
C, F
YUNASKO 330F
20
0.22
1.35
337
40
0.24
1.81
335
60
0.20
0.81
333
80
0.22
0.78
331
100
0.22
0.73
330
20
0.11
0.88
1231
40
0.14
0.47
1224
60
0.16
0.30
1224
80
0.15
0.21
1224
100
0.15
0.16
1223
20
0.28
1.30
1616
40
0.26
0.81
1600
60
0.30
0.66
1593
80
0.30
0.61
1587
100
0.21
0.56
1585
YUNASKO 1230F
Maxwell 1500F
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
25
5.2 Constant power cycling
Fig. 12, 13 demonstrate typical constant power cycling curves for Yunasko 330 F supercapacitor and
Yunasko H3-6000 F hybrid device.
Fig.12. Constant power cycling curves for Yunasko 330 F supercapacitor.
Fig.13. Constant power cycling curves for Yunasko H3-6000 F hybrid device.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
26
Constant power cycling data can be used for Ragone plots, some samples of which are shown in Fig.
14.
Fig. 14. Ragone plot for hybrid devices: 1 – JSR 2000F (ITS test results), 2 – Yunasko H1-8000F (ITS
test results), 3 – Yunasko H3-6000F (Yunasko test results)
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
27
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
28
6. CONCLUSIONS
Analysis of the international practical standards for testing carbon-carbon and hybrid
supercapacitors (USABC Manual, IEC 62391, IEC 62576, testing procedures used by ITS, UC Davis and
others) has been conducted and compared to the standard procedures utilized by YUNASKO
company. The optimal testing procedures have been formulated to be used by the partners of
Energy Caps project.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
29
7. REFERENCES
1. IEC 62391-2 . Fixed electric double layer capacitors for use in electronic equipment. Part 2.
Sectional specification – Electric double layer capacitors for power application.
2. IEC 62576. Electric double layer capacitors for use in hybrid electric vehicles – Test methods
for electrical characteristics.
3. FreedomCAR Ultracapacitor Test Manual. Idaho National Laboratory Report DOE/NE-ID11173, September 21, 2004.
4. A. Burke, M. Miller. Testing of Electrochemical Capacitors: Capacitance, Resistance, Energy
Density, and Power Capability. ISEE’Cap09 Conference, Nantes, 2009.
5. A. Burke, J.R. Miller. Electric Vehicle Capacitor Test Procedures Manual. Idaho National
Engineering Laboratory Report DOE/ID-10491, October 1994
6. S. Zhao, F. Wu, L. Yang, L. Gao, A. Burke. A measurement method for determination of dc
internal resistance of batteries and supercapacitors. Electrochemistry Communications,
2010, v.12, p.242-245.
FP7-IAPP Energy Caps EU funded project - Grant Agreement N° 286210 - All rights reserved.
30