0% found this document useful (0 votes)
23 views6 pages

Capacitor Leakage An

This application note details the measurement of capacitor leakage using a Model 6517A Electrometer, highlighting the importance of insulation resistance and the impact of dielectric types on leakage measurements. It discusses optimal testing methods, including the use of series resistors to mitigate noise and the implementation of automated test systems for efficiency. Additionally, it emphasizes safety precautions necessary when working with high voltage and power levels in testing environments.

Uploaded by

angelo wolf
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
23 views6 pages

Capacitor Leakage An

This application note details the measurement of capacitor leakage using a Model 6517A Electrometer, highlighting the importance of insulation resistance and the impact of dielectric types on leakage measurements. It discusses optimal testing methods, including the use of series resistors to mitigate noise and the implementation of automated test systems for efficiency. Additionally, it emphasizes safety precautions necessary when working with high voltage and power levels in testing environments.

Uploaded by

angelo wolf
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Number 315

Application Note Capacitor Leakage Measurements


Series Using a Model 6517A Electrometer

Introduction possible to im-prove noise, yet small enough to avoid introducing


errors in the final result. This trade-off requires knowing the
Capacitors are very important in all areas of electronics. approximate value of the insulation resistance, even before the
From timing circuits to sample and hold applications, we depend measurement has been made. Even then, finding the optimum
on capacitors to act in a nearly ideal fashion. In many cases, value for the series resistor can be difficult, especially when a
however, complex electrochemical interactions cause capacitors variety of capacitors are to be tested.
to fall short of perfect.
Second, the resistor adds noise to the measurement.
One of the less ideal properties that a capacitor has is Johnson noise is thermal noise created by any resistor. At room
leakage, or insulation resistance (IR). For a given dielectric temperature this is roughly 6.5 × 10–10 ∆f/R amps, p-p. The
material, the effective parallel resistance is inversely proportional current noise in a 1TΩ feedback resistor at a typical 3Hz band-
to the capacitance. This is because the resistance is proportional width would be ~8×10-16A. When measuring an insulation resist-
to the thickness of the dielectric, and inverse to the capacitive ance of 1016Ω at 10V, the noise current will be 80% of the
area. The capacitance is proportional to the area and inverse to the measured current.
separation. Thus, a common unit for qualification of capacitor
leakage is the product of its capacitance and leakage resistance, RF
RL
usually in megohm-microfarads (MΩ·µF).
RS

Dielectric Types V
C
+
For polymer dielectrics such as polystyrene, polycarbonate Test Feedback
or Teflon®, the insulation resistance can range from 104MΩ·µF to Circuit Electrometer
108MΩ·µF, depending on the materials and their purity. For
RF
example, a 1000pF Teflon cap with insulation resistance greater
Gain: 1 + RL
than 1017Ω is specified as >108MΩ·µF. For various ceramics such + RS
1 + R L jω C
as X7R or NPO, insulation resistance can range from 103MΩ·µF
to 106MΩ·µF. Electrolytic capacitors such as tantalum or alumi-
Gain at 3Hz
num have much lower leakage resistances, in the region from without RS : 1 + R F j ωC (Typically 104 - 108)
1MΩ·µF to 100MΩ·µF. For example, a 4.7µF aluminum cap
specified as 50MΩ·µF is guaranteed to have at least 10.6MΩ Gain at 3Hz RF
insulation resistance. with RS : 1 + (Typically 104 - 106)
RS

Figure 1.
Typical Test Method
When measuring extremely low leakage capacitors, there
are a number of things to keep in mind. Normally, a feedback
Alternate Test Circuit
electrometer would be used as shown in Figure 1. The series Even better results will be obtained with the circuit of
resistor (RS) in the measurement loop is necessitated by noise Figure 2. The added resistance of diode DS allows RS to be
considerations. Without the resistor, an electrometer in this reduced to approximately 100kΩ. At the beginning of the
configuration would have a very high noise gain at high frequen- capacitor charge cycle, the current through the diode is relatively
cies. This noise amplification is unacceptable. The series resistor high and the equivalent resistance of the diode is low, allowing
limits the AC noise to a maximum level, although it does make the capacitor to charge quickly. As the capacitor be-comes
the measurement more complex. Since RS appears in the denomi- charged, the current will decrease steadily and the diode resis-
nator of the simplified gain equation for 3Hz, a larger resistance tance will increase, limiting the noise amplification automatically.
decreases the AC noise gain. To make a measurement to 0.1%, Typically, the diode can be a small-signal diode, such as the
the series resistor must be less than 0.1% of the insulation 1N914 or the 1N3595. Note some series resistance is still
resistance to be measured. Thus, the resistor should be as large as required to prevent overload in case the capacitor is shorted.
The diode should be enclosed in a light-tight metal enclo- sequences involving required environmental conditions, the
sure to eliminate photo-electric as well as electrostatic interfer- 6517A has the ability to simultaneously monitor tempera-
ence. ture and humidity. This provides a record of conditions, and
allows for easier determination of temperature coefficients.
RL2
Automatic time-stamping of readings provides a further
record of time-resolved measurements.
RL1
3. An optional switch card will allow repeated testing of up to
CF 10 capacitors to facilitate small batch testing for the labora-
tory environment.
V C
+ 4. The V/I mode simplifies the task of making a single
Test Feedback measurement. In this mode a resistance reading of V/I is
Circuit Electrometer displayed, using its output voltage and measured current
reading. When setting up a measurement system, it is
convenient to have readings in a form that requires no
jωC + 1/RL1
Gain: 1 + interpretation. This is helpful when using the instrument
jωC + 1/R
F L2
without a computer interface.
C
Gain at 3Hz: 1 + (Typically 1 - 1000)
CF
Test Circuits
Figure 2.
For statistical purposes, a quantity of capacitors must be
Figure 3 show some typical response curves using different tested to produce useful data. Obviously, it is impractical to
series resistors. perform these tests manually, so some sort of automated test
system is required. Figure 4 illustrates such a system, which
employs a Model 6517A Electrometer/Source, Model 7158 Low
Current Scanner Cards, and Model 7169A Form C Switch Cards.
The cards must be installed in a switching mainframe, such as a
Model 7002. A computer controls the instruments to perform the
tests automatically.

7169A Form C Switch Card 7158 Low Current Card


R C

Figure 3. Typical response curves


R C

Implementation
The 6517A Electrometer offers several advantages when R C
measuring capacitor leakage.
1. The 6517A contains a low noise, variable 1kV voltage
source for making high resistance measurements, with
built-in current limiting. For a given capacitor, a larger
applied voltage within the voltage rating of the capacitor, LO LO
Model 6517A
will give a larger leakage current. Measuring a larger Voltage Source Output
Electrometer Picoammeter Input

current with the same intrinsic noise floor produces a HI HI


greater signal-to-noise ratio, and a more accurate reading.
Figure 4. Capacitor leakage test system
For capacitors with high voltage ratings, a 1000V source is
supplied. Because the voltage source is continuously In this test system, a single instrument, the Model 6517A,
variable, voltage coefficients are easily obtained. provides both the voltage sourcing and low current measurement
functions. This instrument is particularly useful for this applica-
2. Temperature and humidity can have a significant effect on
tion because it can display either resistance or leakage current and
high resistance measurements. It therefore becomes impor-
will source up to 1000V DC. Due to the limitation of the Model
tant to regulate and measure these quantities. For test
7169A card, the amount of voltage sourced should not exceed
8002-ILC-3 Interlock Cable
A) Connections
WARNING:NO INTERNAL OPERATOR SERVICABLE PARTS,SERVICE BY QUALIFIED PERSONN

LO HI
8002A HIGH RESISTANCE TEST FIXTURE

! PREAMP OUT COMMON V SOURCE


250V PEAK
INPUT
250V PEAK

OUT
!

7056
INTERLOCK

7078-TRX-3 Triax Cable CAUTION:FOR CONTINUED PROTECTION AGAINST FIRE HAZARD,REPLACE FUSE WITH SAME TYPE

Model 6517A

Note: Set fixture mode switch 8607 Banana Plug Cables


to picoammeter operation.

Warning: Connect of fixture


to safety earth ground using
R2
safety ground wire (supplied
with 8002A test fixture).

B) Equivalent Circuit 1 2 3
Input
Amplifier
LCR
Guard Input
Plate Amplifier
HI
RF
pA

LO –
+ To A/D
+ Converter
RX GND
S
S
R1
HI

V-Source Out
V Source
LO
7158
To Interlock
Detection
Lid Interlock Interlock
Circuits

S
Model 8002A
Test Fixture
PREAMP OUTPUT

COMMON 1Ω Normally closed relays disabled


2V ANALOG OUTPUT

Figure 5. Connections for resistance measurements using Model 8002A Figure 6. Capacitance and IR measurement system
test fixture

500V. If the maximum test voltage is only 110V, the 7169A card currents that can become a significant portion of a high resistance
can be replaced with the Model 7111 Form C Switch Card. measurement if not controlled.
One set of switches is used to apply the test voltage to each Connections to the 8002A from the 6517A are shown in
capacitor in turn; a second set of switches connects each capacitor Figure 5. Note that the 8002A requires the use of a 8002-ILC3
to the picoammeter after a suitable soak period. This system tests Interlocking Cable. The interlock feature of the 6517A and
up to ten capacitors but is easily expanded to any reasonable 8002A prevents voltage from the 6517A from being applied when
number. the lid of the 8002A is open. However, once a capacitor has been
charged to a high voltage, the capacitor should be discharged
After the capacitors have been tested, the voltage source
prior to any handling and removal from the fixture in order to
should be set to zero and some time allowed to discharge the
prevent any electrical shock hazards.
capacitors before they are removed from the fixture. Note that in
Figure 4 the capacitors have a discharge path through the More complex test systems are possible, combining the
normally closed contacts of the relays. leakage measurement with capacitance measurements, dielectric
absorption and other tests, if desired. A simplified schematic of
To prevent electric shock, test connections must be config-
such a test system using an LCZ bridge and a picoammeter with a
ured such that the user cannot come in contact with the conduc-
voltage source is shown in Figure 6.
tors, connections, or the DUT. Safe installation requires proper
shielding, barriers, and grounding to prevent contact with
conductors. Example Program and Description
For single capacitor testing, the Model 8002A High An example computer program is shown in Figure 7. This
Resistance Test Fixture can be used with the 6517A. The Model example program illustrates the programming of the 6517A using
8002A has been specifically designed to minimize leakage the built-in test sequence, Capacitor Leakage Test, of the 6517A
that provides a capacitor leakage measurement.
Refer to the program listing for the following program Although this example was written with using BASIC in
description. mind, the program can be modified for Visual Basic, C,
TestPoint, LabVIEW, or any other programming environment
After the 6517A has been cleared, the instrument is
that supports GPIB communication with instruments. Examples
configured for the following functions:
in TestPoint and LabVIEW are available from Keithley at
• Math calculations turned off. www.keithley.com.
• 6517A set to SRQ when buffer is full.
• Retum ASCII data for the leakage current reading and the
Test System Safety
Voltage Source value. Many electrical test systems or instruments are capable of
measuring or sourcing hazardous voltage and power levels. It is
• Set the 6517A to measure current.
also possible, under single fault conditions (e.g., a programming
• Set the 6517A for autoranging. error or an instrument failure), to output hazardous levels even
when the system indicates no hazard is present.
Once these commands are issued, the program checks to
see if the operation has been completed prior to sending any These high voltage and power levels make it essential to
further commands. This gives the 65l7A the appropriate amount protect operators from any of these hazards at all times. Protec-
of time to be configured. tion methods include:

After the initial setup, the operator is then prompted for the • Design test fixtures to prevent operator contact with any
desired soak voltage, the soak time, and the discharge time. The hazardous circuit.
discharge time allows the capacitor to be discharged so that it • Make sure the device under test is fully enclosed to
may be handled. With these values entered, the program then protect the operator from any flying debris.
sends the test sequence commands to the 6517A with the
• Double insulate all electrical connections that an operator
prompted values for the test. Again, once these commands are
could touch. Double insulation ensures the operator is
issued, the program checks to see if the operation has been
still protected, even if one insulation layer fails.
completed prior to sending any further commands. This gives the
6517A the appropriate amount of time to be configured. • Use high-reliability, fail-safe interlock switches to
disconnect power sources when a test fixture cover is
When the operation complete function has been satisfied,
opened.
the program then arms the test sequence to begin the test. During
the testing phase, the computer program waits for the SRQ to be • Where possible, use automated handlers so operators do
identified when the data buffer has been filled. When the SRQ not require access to the inside of the test fixture or have
has been identified, the program reads the measurement status a need to open guards.
register. This action clears the SRQ from the status register. • Provide proper training to all users of the system so they
Clearing the SRQ is important, especially if the test is run again. understand all potential hazards and know how to protect
In addition to clearing the status register, getting the data from the themselves from injury.
buffer clears the event that caused the SRQ.
It is the responsibility of the test system designers, integra-
The data is obtained and placed into the variable CapLeak$. tors, and installers to make sure operator and maintenance
Because the data is returned in ASCII format and there are two personnel protection is in place and effective.
values associated with the single capacitor leakage current value,
the string must be parsed in order to separate the current reading
and the voltage source values. As shown in the example, BASIC
uses the MIDS$ function to obtain a subset ofthe string.
Figure 7: Software Example Listing

Sub SetupInst()

'Send Reset command to 6517A. 6517A at Address 27.

Call Send(27, "*RST", status%)

'Set 6517A initial measurement functions, range and measurement elements.

Call Send(27, "*CLS", status%) 'Clear all status registers


Call Send(27, ":Calc1:Stat Off",status%) 'Disable Math calculations
Call Send(27, ":Stat:Meas:Enab 512;*SRE 1", status%) 'Set to SRQ on Buffer Full,
Call Send(27, ":Form:Data ASC;Elem Read,Vso", status%) 'Return readings as ASCII values
'floating point format and only return the
'readings and the Vsource value.
Call Send(27, ":Sens:Func 'Curr", status%) 'Set measurement mode to Current
Call Send(27, ":Sens:Curr:Rang:Auto On", status%) 'Set for autoranging
Call Send(27, "*Opc?”, status%) 'Query for Operation Complete
Call Enter(OpCom$, 20, 1%, 27, status%) 'Get Operation Complete Bit

End Sub

Sub TakeReading()

Input "Input Soak Voltage" SoakVoltage 'Enter Soak Voltage


Input "Input Soak Time"; SoakTime 'Enter Soak Time
Input "Input Discharge Time"; DischargeTime 'Enter Discharge Time

'Enter values into Test Sequence Commands and run test

Call Send(27, ":Trace:Clear", status%) 'Clear contents of trace buffer


Call Send(27, ":Tseq:Type CLE", status%) 'Set test sequence for Cap Leakage
Call Send(27, ":Tseq:CLE:Svol " +str$(SoakVoltage), status %)
'Set Soak Voltage for sequence
Call Send(27, ":Tseq:CLE:Stime "+str$(SoakTime), status%) 'Set Soak Time for sequence
Call Send(27, ":Tseq:CLE:Dtime "+str$(DischargeTime), status%)
'Set Discharge Time for sequence

Call Send(27, "Tseq:Tso Imm", status%) 'Enable Immediate start of sequence


Call Send(27, "*Opc?", status%) 'Query for Operation Complete
Call Enter(OpCom$, 20, 1%, 27, status%) 'Get Operation Complete Bit

'Arm and start sequence


Call Send(27, ":Tseq:Arm", status%) 'Start Seqeunce

'At this point, the program waits for the SRQ from the 6517A when the data buffer is full.

WaitSRQ:

IF (NOT (srq%)) THEN GOTO WaitSRQ

'When the SRQ is identified, execute the following code:

Call Send(27, ":Stat:Meas:Event?", status%) 'Get status and clear status


Call Enter(poll$, 50, 1%, 27, status%) 'Get poll value

Call Send(27, ":Trace:Data?", status%) 'Query for buffered readings


Call Enter(CapLeak$, 30, 1%, 27, status%) 'Get Reading and Vsource

'Since there are two values returned, the reading and the Vsource value, the CapLeak$
'must be parsed out to get the two values. The data sent back to the host computer will
'be in this format:
'
' (+/-)1234.567E(+/-)00,(+/-)0010.000(LF)
'
'Note that (+/-) represents one character. Thus the 6517A returns a 'total of 24 characters
'including the (,) comma and the (LF) Line Feed character.

PRINT "Leakage Current Value=";MID$(CapLeak$, 1, 13) 'Extract 13 characters for the leakage value
PRINT "Soak Voltage=";MID$(CapLeak$, 15, 13) 'Extract next 13 characters after comma for
'VSource value.

End Sub
Specifications are subject to change without notice.
All Keithley trademarks and trade names are the property of Keithley Instruments, Inc.
All other trademarks and trade names are the property of their respective companies.

Keithley Instruments, Inc. 28775 Aurora Road • Cleveland, Ohio 44139 • 440-248-0400 • Fax: 440-248-6168
1-888-KEITHLEY (534-8453) www.keithley.com
BELGIUM: Keithley Instruments B.V. Bergensesteenweg 709 • B-1600 Sint-Pieters-Leeuw • 02-363 00 40 • Fax: 02/363 00 64
CHINA: Keithley Instruments China Yuan Chen Xin Building, Room 705 • 12 Yumin Road, Dewai, Madian • Beijing 100029 • 8610-6202-2886 • Fax: 8610-6202-2892
FRANCE: Keithley Instruments Sarl 3, allée des Garays • 91127 Palaiseau Cédex • 01-64 53 20 20 • Fax: 01-60 11 77 26
GERMANY: Keithley Instruments GmbH Landsberger Strasse 65 • 82110 Germering • 089/84 93 07-40 • Fax: 089/84 93 07-34
GREAT BRITAIN: Keithley Instruments Ltd. Unit 2 Commerce Park, Brunel Road • Theale • Reading • Berkshire RG7 4AB • 0118 929 7500 • Fax: 0118 929 7519
INDIA: Keithley Instruments GmbH Flat 2B, Willocrissa • 14, Rest House Crescent • Bangalore 560 001 • 91-80-509-1320/21 • Fax: 91-80-509-1322
ITALY: Keithley Instruments s.r.l. Viale San Gimignano, 38 • 20146 Milano • 02-48 39 16 01 • Fax: 02-48 30 22 74
KOREA: Keithley Instruments Korea 2FL., URI Building • 2-14 Yangjae-Dong • Seocho-Gu, Seoul 137-130 • 82-2-574-7778 • Fax: 82-2-574-7838
NETHERLANDS: Keithley Instruments B.V. Postbus 559 • 4200 AN Gorinchem • 0183-635333 • Fax: 0183-630821
SWITZERLAND: Keithley Instruments SA Kriesbachstrasse 4 • 8600 Dübendorf • 01-821 94 44 • Fax: 01-820 30 81
TAIWAN: Keithley Instruments Taiwan 1FL., 85 Po Ai Street • Hsinchu, Taiwan, R.O.C. • 886-3-572-9077• Fax: 886-3-572-9031

© Copyright 2001 Keithley Instruments, Inc. No. 1682


Printed in the U.S.A. 8012KDCI

You might also like