-
SYS MENU MEASDISPFUNC : Z-
FREQ : 100.000 kHzLEVEL : 1.00 V
RANGE : AUTOBIAS : 0.000 VINTEG : LONG BIN
No.
BINCOUNT
LISTSWEEP
Vm : 50.00mVCORR: OPEN, SHORT
Im : 10.74mA
150.000 Ω:Z -90.000 deg:θ
θ
Agilent8 Hints for Successful Impedance MeasurementsApplication
Note 346-4
Characterizing Electronic Componentsto Achieve Designed Circuit
Performance
-
Contents
HINT 1. Impedance Parameters
HINT 2. Measurements Depend on Test Conditions
HINT 3. Choose Appropriate Instrument Display Parameter
HINT 4. A Measurement Technique has Limitations
HINT 5. Perform Calibration
HINT 6. Perform Compensation
HINT 7. Understanding Phase Shift and Port Extension Effects
HINT 8. Fixture and Connector Care
ImpedanceMeasurements forEngineersImpedance is measured using a
vari-ety of techniques. A particular tech-nique is selected
according to thetest frequency, the impedanceparameter to be
measured and thepreferred display parameters.
The Auto-Balancing Bridgetechnique is exceptionally accurateover
a broad impedance range (mΩto the order of 100 MΩ). The fre-quency
range this technique can be applied to is from a few Hz to 110
MHz.
The IV and RF-IV techniques arealso very accurate over a
broadimpedance range (mΩ to MΩ).These techniques can be appliedfrom
40 Hz to 3 GHz.
The Transmission/Reflectiontechnique is applied over the
broadest frequency range (5 Hz to110 GHz). This technique
deliversexceptional accuracy near 50 Ω or75 Ω.
LCR meters and impedance analyz-ers are differentiated primarily
bydisplay properties. An LCR meterdisplays numeric data, while
animpedance analyzer can display datain either numeric or graphic
formats.
Figure 0-1. Accuracy Profile
The techniques employed by theseinstruments are independent of
ana-lyzer type, and can be RF-IV, IV orAuto-Balancing Bridge
(dependingon frequency).
Engineers perform impedance meas-urements for a variety of
reasons. Ina typical application, an electroniccomponent used in a
new circuitdesign is characterized. Normally,component
manufacturers state onlynominal impedance values.
Design decisions, as well as deci-sions affecting the production
of theassembled product, depend to somedegree on the impedance
valuesattributed to the product’s compo-nents. The performance and
qualityof the final product are thereforedetermined in part by the
accuracyand thoroughness with which itscomponents are
characterized.
This application note provides help-ful information for using
the Auto-Balancing Bridge, IV and RF-IV tech-niques. Refer to
Agilent ApplicationNote 1291-1, 8 Hints for MakingBetter Network
AnalyzerMeasurements (literature number 5965-8166E) for information
on theTransmission/Reflection technique.
2
-
HINT 1.ImpedanceParametersImpedance is a parameter used
toevaluate the characteristics of elec-tronic components. Impedance
(Z)is defined as the total opposition acomponent offers to the flow
of analternating current (AC) at a givenfrequency.
Impedance is represented as a com-plex, vector quantity. A polar
coordi-nate system is used to map thevector, where quadrants one
andtwo correspond respectively to pas-sive inductance and passive
capaci-tance. Quadrants three and fourcorrespond to negative
resistance.The impedance vector consists of areal part, resistance
(R), and animaginary part, reactance (X).
Figure 1-1 shows the impedancevector mapped in quadrant one
ofthe polar coordinate system.
Capacitance (C) and inductance (L)are derived from resistance
(R) andreactance (X). The two forms ofreactance are inductive (XL)
andcapacitive (XC).
The Quality Factor (Q) and theDissipation Factor (D) are
alsoderived from resistance and reac-tance. These parameters serve
asmeasures of reactance purity. WhenQ is larger or D is smaller,
the quali-ty is better. Q is defined as the ratioof the energy
stored in a componentto the energy dissipated by the com-ponent. D
is the inverse of Q. D isalso equal to “tan δ”, where δ is
thedielectric loss angle (δ is the com-plementary angle to θ, the
phaseangle). Both D and Q are dimension-less quantities.
Figure 1-2 describes the relationshipbetween impedance and
thesederived parameters.
Q =
D =
Capacitor
Inductor
RX L
X C
R
R
Z
ZjX L
jX C
-j
+j
R jX L
R - jX C
X L = 2πfL = ωL
X C =
=
1
ωC
2πfC1
δ
δ
θ
θ
Figure 1-1. Impedance Vector Figure 1-2. Capacitor and Inductor
Parameters
3
-
HINT 2.MeasurementsDepend on TestConditionsA manufacturer’s
stated impedancevalues represent the performance ofa component
under specific testconditions, as well as the tolerancepermitted
during manufacture.When circuit performance requiresmore accurate
characterization of acomponent, it is necessary to verifythe stated
values, or to evaluatecomponent performance under oper-ating
conditions (usually differentthan the manufacturer’s test
condi-tions).
Frequency dependency is commonto all components because of
para-sitic inductance, capacitance andresistance.
Figure 2-1 describes ideal and para-sitic frequency
characteristics of atypical capacitor.
Figure 2-1. Frequency Characteristics of a Capacitor
Signal level (AC) dependency isexhibited in the following ways
(see Figure 2-2):
• Capacitance is dependent on AC voltage level (dielectric
constant, K, of the substrate).
• Inductance is dependent on AC current level (electromagnetic
hysteresis of the core material).
The AC voltage across the compo-nent can be derived from the
com-ponent’s impedance, the sourceresistance, and the signal
sourceoutput (Figure 2-3).
An automatic level control (ALC)function maintains a constant
volt-age across the DUT (device undertest). It is possible to write
an ALCprogram for instruments that have alevel monitor function,
but not abuilt-in ALC.
Control of measurement integrationtime allows reduction of
unwantedsignals. The averaging function isused to reduce the
effects of randomnoise. Increasing the integrationtime or averaging
allows improvedprecision, but with slower measure-ment speed.
Detailed explanations ofthese test parameters can be foundin the
instrument operating manuals.
Other physical and electrical factorsthat affect measurement
resultsinclude DC bias, temperature,humidity, magnetic fields,
light,atmosphere, vibration, and time.
(a) AC Voltage Level Dependency ofCeramic Capacitor
Measurement
(b) AC Current Level Dependency ofCore Inductor Measurement
C
0
AC Test Voltage
Dielectric Constant High Dielectric
Constant Medium
Dielectric Constant Low
(No Dependency)
AC Test Current
L
0(No Dependency)
˜
R X
DUT
V dut
SignalSource
Source Resistance
Rs
V A
V dut = Vo *R2 + X2
VoRs
DUT
Feedback
(Rs+R)2 + X2
Vm
Vo Idut
A
Figure 2-2. Signal Level Dependency Figure 2-3. Applied Signal
and Constant Level Mechanism
4
-
HINT 3.Choose AppropriateInstrument DisplayParameterMany modern
impedance measuringinstruments measure the real andthe imaginary
parts of an impedancevector and then convert them intothe desired
parameters.
When a measurement is displayed asimpedance (Z) and phase (θ),
theprimary element (R, C, or L) as wellas any parasitics are all
representedin the |Z| and θ data.
When parameters other than imped-ance and phase angle are
displayed,a two-element model of the compo-nent is used. These
two-elementmodels are based on a series or par-allel circuit mode
(Figure 3-1), andare distinguished by the subscripts“p” for
parallel and “s” for series (Rp,Rs, Cp, Cs, Lp, or Ls).
No circuit components are purelyresistive or purely reactive. A
typicalcomponent contains many parasiticelements. With the
combination ofprimary and parasitic elements, acomponent acts like
a complex cir-cuit.
Recent, advanced impedance analyz-ers have an Equivalent
CircuitAnalysis Function that allows analy-sis of the measurement
result in theform of three- or four-element cir-cuit models (Figure
3-2). Use of thisfunction enables a more completecharacterization
of a component'scomplex residual elements.
Figure 3-1. Measurement Circuit Mode Figure 3-2. Equivalent
Circuit Analysis Function
5
-
HINT 4.A MeasurementTechnique HasLimitationsThe most frequently
asked questionin engineering and manufacturing isprobably: “How
accurate is thedata?”
Instrument accuracy depends on theimpedance values being
measured,as well as the measurement technol-ogy employed (see
Figure 0-1).
To determine the accuracy of ameasurement, compare the meas-ured
impedance value of the DUT tothe instrument accuracy for
theapplicable test conditions.
Figure 4-1 shows that a 1 nF capaci-tor measured at 1 MHz
exhibits animpedance of 159 Ω.
Instrument accuracy specificationsfor D or Q measurements are
usuallydifferent than accuracy specifica-tions for other impedance
terms.
In the case of a low-loss (low-D/high-Q) component, the R-value
isvery small relative to the X-value.Small changes in R result in
largechanges in Q (Figure 4-2).
The measurement error is on theorder of the measured R-value.
Thiscan result in negative D or Q values.
Be aware that measurement errorincludes error introduced by
theinstrument as well as by the test fix-ture.
159
1 10 100 1K 10K 100K 1M10M 100M 1G
10M
1M
100K
10K
1K
100
10
1100m
1mF10mF100mF
100uF
10uF
1uF
100nF
10nF 1nF10pF
100fF 1fF
100pF 1pF10fF
Impe
danc
e (Ω
)
1 nF at 1 MHz
Impedance value of capacitor
Z =
(ω = 2πf)
Test frequency (Hz)
1ωC
Figure 4-1. Capacitor Impedance and Test Frequency Figure 4-2.
Concept of the Q Error
6
-
HINT 5.Perform CalibrationCalibration is performed to define
areference plane where the measure-ment accuracy is specified
(Figure5-1). Normally, calibration is per-formed at the
instrument's test port.Corrections to raw data are based
oncalibration data.
A baseline calibration is performedat service centers for
Auto-BalancingBridge instruments such that thespecified accuracy
can be realizedfor a period of time (usually twelvemonths)
regardless of the instru-ment settings. With these instru-ments,
operators do not requirecalibration standards.
Baseline calibration for non-Auto-Balancing Bridge
instrumentsrequires that a set of calibrationstandards be used
after instrumentinitialization and setup. This hintprovides
information that may behelpful when using calibration stan-dards to
establish calibration forthese instruments.
Some instruments offer the choice offixed-mode or user-mode
calibration.Fixed-mode calibration measurescalibration standards at
predeter-mined (fixed) frequencies. Calibra-tion data for
frequencies betweenthe fixed, calibrated points are
inter-polated.
Fixed-mode calibration sometimesresults in interpolation errors
at fre-quencies between the fixed, calibrat-ed points. At higher
frequencies,these errors can be substantial.
User-mode calibration measures cali-bration standards at the
same fre-quency points the user has selectedfor a particular
measurement. Therecan be no interpolation errors asso-ciated with
user-mode calibration.
It is important to recognize that theoperator-established
calibration isvalid only for the test conditions(instrument state)
under which cali-bration is performed.
Calibration Defines a Reference Plane atWhich Measurement
Accuracy is Specified.
Calibration Plane
Z Analyzer
Z?
Calibration Standards
Short Open LoadLow-LossCapacitor*
OS0Ω –90o50Ω
*Agilent 4291B, 4286A and 4287A
Figure 5-1. Calibration Plane
7
-
HINT 6.PerformCompensationCompensation is not the same as
calibration. The effectiveness ofcompensation depends on
theinstrument calibration accuracy,therefore compensation must
beperformed after calibration has beencompleted.
When a device is directly connectedto the calibration plane, the
instru-ment can measure within a specifiedmeasurement accuracy.
Since a testfixture or adapter is usually connect-ed between the
calibration plane andthe device, the residual impedanceof the
interface must be compensat-ed for to perform accurate
measure-ments.
Additional measurement error intro-duced by a test fixture or
adaptercan be substantial. The total meas-urement accuracy consists
of theinstrument accuracy plus the errorfrom sources that exist
between theDUT and the calibration plane.
It is important to verify that errorcompensation is working
properly. Ingeneral, the impedance value for anopen condition
should be greaterthan 100 times the impedance of theDUT. In
general, the impedancevalue for a short condition should beless
than 1/100 the impedance of theDUT.
Open compensation reduces or elim-inates stray capacitance,
while shortcompensation reduces or eliminatesthe unwanted
resistance and induc-tance of fixturing.
When performing an open or a shortmeasurement, keep the
distancebetween the UNKNOWN terminalsthe same as when the DUT is
contacted. This keeps parasiticimpedance the same as when
meas-urements are performed.
Perform load compensation whenthe measurement port is extended
anon-standard distance, the configu-ration uses additional passive
cir-cuits/components (for example, abalun, attenuator, or filter),
or whena scanner is used. The impedancevalue of the load must be
accuratelyknown. A load should be selectedthat is similar in
impedance (at alltest conditions) and form-factor tothe DUT. Use a
stable resistor orcapacitor as the LOAD device.
It is practical to measure a loadusing open/short compensation
anda non-extended fixture to determinethe load impedance. The
valuesmeasured can then be input as com-pensation standard
values.
Figure 6-1. OPEN/SHORT Compensation
8
-
HINT 7.Understanding PhaseShift and PortExtension EffectsCable
length correction, port exten-sion, or electrical delay is used
toextend or rotate the calibrationplane to the end of a cable or
thesurface of a fixture. This correctionreduces or eliminates phase
shifterror in the measurement circuit.
When the measurement port isextended away from the
calibrationplane (Figure 7-1), the electricalcharacteristics of the
cable affect thetotal measurement performance. Toreduce the
resulting effects:
• Make measurement cables as short as possible.
• Use well-shielded, coaxial cables to prevent influence from
external noise.
• Use low-loss coaxial cables to keep from degrading accuracy,
since the port extension method assumes lossless cable.
A phase-shift-induced error occursdue to the test fixture, which
cannot be reduced using OPEN/ SHORTcompensation.
When working in the RF region, cali-bration should be performed
at theend of an extension cable. If calibra-tion standards cannot
be inserted,port extension can be used in thisregion for short and
well-character-ized distances.
When using the Auto-BalancingBridge technique with
non-standardcables or extensions, open/short/loadcompensation
should be performedat the terminus of an extension orfixture.
Auto-Balancing Bridge prod-ucts use cable length compensationfor
standardized test cables (1, 2, or4 meters). At the terminus of
thestandardized extension cable, theshields should normally be
connect-ed together.
Port extension in any form has limi-tations. Since any extension
willcontribute to losses in the measure-ment circuit and/or phase
error, it isimperative that the limitations of themeasurement
technique be fullyunderstood prior to configuring anextension.
ImpedanceMeasuringInstrument
Extension Cable TestFixture
Device Under Test
Figure 7-1. Measurement Port Extension
9
-
HINT 8.Fixture andConnector CareHigh-quality electrical
connectionsinsure the capability to make precisemeasurements. At
every connection,the characteristics of the mating sur-faces vary
with the quality of con-nection. An impedance mismatch atmating
surfaces will influence propa-gation of the test signal.
Attention should be paid to the mat-ing surfaces of test ports,
adapters,calibration standards, fixture con-nectors, and test
fixtures. The quali-ty of connections depends on thefollowing:
• composition• technique• maintenance• cleanliness• storage
CompositionIt has been said that a chain is asstrong as the
weakest link. The sameis true for a measurement system.
Iflow-quality cables, adapters or fix-tures are used in a test
system, theoverall quality of the system isreduced to that of the
lowest-qualityinterface.
TechniqueThe use of a torque wrench andcommon sense insures that
damagedoes not occur when making repeat-ed connections. Damage
includesscratching and deformation of themating surfaces.
MaintenanceMany mating surfaces are designedto allow for the
replacement of partsthat degrade with use. If a matingsurface
cannot be repaired, regularlyscheduled replacement is advised.
CleanlinessThe use of non-corrosive/non-destructive solvents
(such as de-ion-ized water and pure isopropylalcohol) and lint-free
wipes insuresthat the impedance at mating sur-faces is not
influenced by residualoils or other impurities. Note thatsome
plastics are denatured with theuse of isopropyl alchohol.
StorageIf a case is not provided with anaccessory, plastic caps
should beused to cover and protect all matingsurfaces when not in
use.
10
-
Agilent Technologies’Impedance ProductLineupAgilent offers the
widest selection ofimpedance measuring equipment foryour
applications. An overview ofthese instruments is given below.For
more information, refer to theproduct literature listed at the
endof this note.
LCR Meters LCR meters can easily and accurate-ly evaluate
components such ascapacitors, inductors, transformersand
electromechanical devices. Theability of these instruments to
applyspecific measurement conditions(such as test frequency and
signallevel) is important in the R&D, pro-duction test and QA
environments.
Impedance Analyzers Agilent impedance analyzers canmeasure
characteristic changes incomponent performance resultingfrom
changes in specific measure-ment conditions. The
characteristicchanges in performance can be dis-played in a
graphical format. Theseanalyzers provide sophisticatedfunctions,
such as markers and
programming, which ease evaluationof measurement results. They
alsohave features that enable character-istic evaluations for
R&D, as well asreliability evaluations (includingtemperature
characteristics) for QApurposes.
Network Analyzers Network analyzers allow impedancemeasurements
at RF and microwavefrequencies using the Transmission/Reflection
technique. Their graphicaldisplays have markers and program-ming
functions that simplify theanalysis of measurement results.Agilent
network analyzers are suit-able for both R & D and QA use.
Combination Analyzers Combination analyzers from Agilentprovide
three capabilities—vectornetwork, spectrum and
impedancemeasurements—in one box. Theseinstruments deliver broad
function-ality to engineers over a wide rangeof applications from
circuit design tocomponent evaluation.
11
-
For more assistance with your test & measurement needs go
to
www.agilent.com/find/assist
Or contact the test and measurementexperts at Agilent
Technologies(During normal business hours)
United States:(tel) 1 800 452 4844
Canada:(tel) 1 877 894 4414(fax) (905) 206 4120
Europe:(tel) (31 20) 547 2000
Japan:(tel) (81) 426 56 7832(fax) (81) 426 56 7840
Latin America:(tel) (305) 267 4245(fax) (305) 267 4286
Australia:(tel) 1 800 629 485 (fax) (61 3) 9272 0749
New Zealand:(tel) 0 800 738 378 (fax) 64 4 495 8950
Asia Pacific:(tel) (852) 3197 7777(fax) (852) 2506 9284
Product specifications and descriptionsin this document subject
to change without notice.
Copyright © 2000Agilent TechnologiesPrinted in USA
06/005968-1947E
Product Literature1. LCR Meters, Impedance
Analyzers, and Test Fixtures Selection Guide, literature number
5952-1430E.
2. RF Economy Network Analyzer, literature number
5967-6316E.
Key Web Resourseswww.agilent.com/find/component_test
12