Agilent 4395A/Agilent 4396BNetwork/Spectrum/ImpedanceAnalyzer

Agilent

Network, Spectrum, and ImpedanceEvaluation of Electronic Circuits andComponentsApplication Note 1308-1

2

IntroductionWith the current trends requiringhigher performance, smaller physicalsize, lower cost, and higher reliability,fast cycle time is an increasinglyimportant part of the product design.This is true not only in the consumerindustry, but also in the communica-tion and data processing industries.

As shown in Figure 1, the develop-ment procedure involves designingthe overall architecture, evaluatingelectronic components, building pro-totypes, and evaluating circuit per-formance. From the design point ofview, it is essential to measure theimpedance characteristics of electron-ic components such as capacitors,

Figure 1. Circuit Development Process

inductors, or other devices at theactual working condition. An imped-ance analyzer makes this task easy.

After making a prototype, a review ofthe circuit operation requires meas-urement of the following parameters:harmonics, noise, transmission char-acteristics, and reflection characteris-tics. The vector network analyzer, andthe spectrum analyzer must be usedfor making these measurements. TheAgilent Technologies 4395A/4396BNetwork/Spectrum/ImpedanceAnalyzer Family combines three ana-lyzer functions in one instrument. This application note describes howthe 4395A/96B can be used to contri-bute fast cycle time for product development.

Step1 Step2 Step3 Step4

Design Circuit CharacterizeComponents Make Circuit Evaluate circuit

perfomance

Inductor

Capacitor

Resonator

Oscillator CircuitFilter/ AMP

CircuitReceiver

Transmitter

NA

SA

3

The Agilent Combination Analyzers have the following featuresThree analyzers in One boxAs the name implies, the Agilent4395A/4396B Network/Spectrum/Impedance Analyzer Family can per-form vector network, spectrum, and(optional) impedance measurements.The combination analyzer family doesnot compromise vector network, spec-trum, or impedance performance. Itis a breakthrough in test instruments,giving you outstanding performanceas a full-capability combination ana-lyzer. Precision measurements andimproved efficiency are possible withminimal training. Compared withusing separate instruments, the4395A/96B will save equipment costand bench space.

Vector Network Analyzer PerformanceThe Agilent Combination AnalyzerFamily offers you fast measurementwith wide dynamic range. Transmis-sion and reflection data can be pro-vided with an optional Reflection/Transmission Test Set or optional S-Parameter Test Set. As a vector net-work analyzer, the 4395A operatesfrom 10 Hz to 500 MHz. The 4396Boperates from 100 kHz to 1.8 GHz.

Spectrum Analyzer PerformanceThe Agilent Combination Analyzers,designed with new digital techniques,outperform the traditional analogspectrum analyzer. Agilent’s Combina-tion Analyzers feature a Fast FourierTransform (FFT) digital-signal pro-cessing (DSP) technique for 20 to 100times faster narrow-band spectrummeasurement, when compared withswept-tuned spectrum analysis. TheAgilent analyzer’s stepped FFT tech-

nique (4395A: all RBWs, 4396B: 1Hz to3kHz RBW) breaks the speed barrierto give you lower noise floors withoutsacrificing speed. In addition, lowphase noise provides improved signalresolution. Option 1D6 (The Time-Gated Spectrum Analysis Function)performs accurate burst signal analy-sis for burst-modulated signal evalua-tion. As a spectrum analyzer, the 4395Aoperates from 10 Hz to 500 MHz. The4396B operates from 2 Hz to 1.8 GHz.

Impedance Analyzer PerformanceWhen equipped with Option 010 andthe Agilent 43961A, CombinationAnalyzers can perform direct imped-ance measurements. Measurementparameters such as Z , θ, C, L, Q, D,and more can be displayed directly on the color display. A built-in lumpedequivalent circuit function aids cir-cuit modeling and simulation. As anImpedance Analyzer, the 4395A oper-ates from 100 kHz to 500 MHz. The4396B operates from 100 kHz to 1.8 GHz.

Other Useful FunctionsIBASIC, a subset of the HT BASICprogramming language, is includedwith the standard 4395A/96B. IBASICis extremely powerful and easy touse. It can be used for automatedtesting, analysis of measurementresults, or control of external equip-ment via GPIB. Files such as instru-ment state files, TIFF files, and datafiles can be transferred via GPIB tothe controller/PC (host), which caneasily manipulate the files. Usabilityand productivity are improved throughfeatures such as the DOS supportedFloppy Disk Drive, the list sweepfunction, the marker function, andthe limit line function.

4

Table 1. Agilent 4395A Major Specification

Network Analyzer Specification Spectrum Analyzer Specification Impedance Analyzer Specification2

Frequency Range 10 Hz to 500 MHz1 Frequency Range 10 Hz to 500 MHz Frequency Range 100 kHz to 500 MHzFrequency Resolution 1 mHz Noise Sidebands <-104 dBc/Hz typical Meas. Parameter Z , θz, R, X, Y , θy,

at 10 kHz offset G, B, Cs, Cp, Ls, Lp, Rp,Output Power Range -50 to 15 dBm Resolution Bandwidth 1 Hz to 1 MHz in Rs, X, D, Q, Γ , Γx ,Γy

1-3-10 steps Z Accuracy 3 % (typical, basic accuracy)Dynamic Range 115 dB@10 Hz IFBW Dynamic Range > 100 dB third-order Source Level -56 dBm to +9 dBm(at DUT)

free dynamic range DC Bias 40 V (20 mA(max))Dynamic Accuracy 0.05 dB/0.3 deg. Level Accuracy 0.8 dB@50 MHz (Opt-001 DC source orCalibration full two-port Sensitivity -145 dBm/Hz External DC source is

@freq. = 10 MHz required.)Compensation open/short/load

Port Extension

Standard Features : Instrument BASIC, GPIB port, 3.5" floppy disk drive, direct print, RAM disk, VGA Monitor OutputOptional Features : Impedance measurement (Opt. 010), Time-Gated spectrum Analysis (Opt. 1D6),

High-Stability Frequency Reference (Opt. 1D5),50 Ω to 75 Ω Spectrum Input Impedance Conversion (Opt. 1D7),DC Source(> 40V, 100 mA (ALC)) (Opt. 001)

1. 100 kHz to 500 MHz if using the 87511A/B S-parameter test set. 2. With Option 010 and the 43961A RF impedance test kit.

Table 2. Agilent 4396B Major Specification

Network Analyzer Specification Spectrum Analyzer Specification Impedance Analyzer Specification 2

Frequency Range 100 kHz to 1.8 GHz1 Frequency Range 2 Hz to 1.8 GHz Frequency Range 100 kHz to 1.8 GHzFrequency Resolution 1 mHz Noise Sidebands <-113 dBc/Hz typical Meas. Parameter Z , θz, R, X, Y , θy,

at 10 kHz offset G, B, Cs, Cp, Ls, Lp, Rp,Output Power Range -60 to 20 dBm Resolution Bandwidth 1 Hz to 3 MHz in Rs, X, D, Q, Γ , ΓX, Γy

1-3-10 steps Meas. range 2 Ω to 5 kΩDynamic Range >120 dB@10 Hz IFBW Dynamic Range > 100 dB third-order Z Accuracy 3% (typical, basic accuracy)

dynamic range Source Level -66 dBm to + 14 dBmDynamic Accuracy 0.05 dB/0.3 deg. Overall Level Accuracy < 1.0 dB (at DUT)Calibration full two-port Sensitivity < -147 dBm/Hz DC Bias 40 V (20 mA(max))

@freq. = 1 GHz (External DC biassource is required.)

Compensation open/short/loadPort Extension

Standard Features : Instrument BASIC, GPIB port, 3.5" floppy disk drive, direct print, RAM disk, VGA Monitor OutputOptional Features : Impedance measurement (Opt. 010), Time-Gated spectrum Analysis (Opt. 1D6),

High-Stability Frequency Reference (Opt. 1D5),50 Ω to 75 Ω Spectrum Input Impedance Conversion (Opt. 1D7)

1. 300 kHz to 1.8 GHz if using the 85046A/B S-parameter test set. 2. With Option 010 and the 43961A RF impedance test kit.

5

Agilent Combination AnalyzerFamily Measurement ExamplesThe Combination Analyzer Family isa powerful tool for effective evalua-tion of electronic circuit and deviceperformance. The following showsyou some examples of measurementsmade by the combination analyzers.

Amplifier EvaluationAmplifier characterization requiresthe evaluation of a variety of meas-urement parameters via vector net-work analysis and spectrum analysis.Figure 2 shows the major measure-ment parameters for amplifierevaluation.

Network Evaluation for AmplifierThe measurement configuration fornetwork analysis of an amplifier isshown in Figure 3. Either an optionalReflection/Transmission test set oran optional S-parameter test set isrequired to perform this analysis.

Gain and Phase MeasurementThe amplifier gain is defined as theratio of the amplifier output power(delivered to a Z0 load) to the inputpower (delivered from a Z0 source),where (Z0) is the characteristicimpedance of the system. The Ampli-fier gain is most commonly specifiedas a typical or minimum value over a specified frequency range, whileassuming that input and output sig-nals are in the amplifier’s linear oper-ating range. Figure 4 shows the gainand phase measurement result of anamplifier.

Agilent 4395A/96B Agilent 4395A/96B

Agilent 87512A R/TTest Set

Amplifierunder test

Amplifierunder test

S-parameterTest Set

(1) R/T Test Set is used (2) S-Parameter Test is used

Figure 3. Measurement Configuration for Amplifier Network Analysis.

vAmplifier Evaluation

SANoiseIMDHarmonicsSpuriousSNR

NAGainPhaseGroup DelayGain CompressionS11 & S22

Figure 2. Major Measurement Parametersof Amplifier

Figure 4. Gain/Phase of the Amplifier

6

Gain Compression MeasurementThe amplifier gain at a single fre-quency is based on ideally linear per-formance between the input powerand the output power. The real ampli-fier gain is nonlinear. The outputpower becomes saturated even if theinput power is increased. The mostcommon measurement of amplifiercompression is the 1-dB compressionpoint. This is defined as the inputpower which results in a 1-dB decreasein amplifier gain. The easiest way tomeasure the 1-dB compression pointis to directly display normalized gain(ratio between the reference channeland the test channel).

Figure 5 shows the gain compressionmeasurement result (normalized gain).The flat part of the trace is linear,and the curved part (of the right side)corresponds to compression causedby higher input power. The AgilentCombination Analyzers have the ability to do power sweeps where themaximum range is 20 dB. The markerfunction can easily indicate the 1-dBdecrease point. These functions helpyou to easily evaluate the gain com-pression of the amplifier.

Return Loss (S11, S22) MeasurementWhen S11, S22, or Return loss aremeasured, either the Reflection/Transmission Test Set or the S-Para-meter test set is required (as shownin Figure 3).

Figure 5. Gain Compression of Amplifier

The input and output in a high fre-quency circuit are matched with acharacteristic impedance, which maydrift from the ideal characteristicimpedance depending on frequency.This change in characteristic imped-ance will cause reflections. The indi-cated reflection parameters are thereflection coefficient (Gamma (S11 orS22)) and return loss(Rl). The returnloss is calculated by using the follow-ing formula.

Rl = -20x log(S11) or -20x log(S22)

When the input and the output areperfectly matched, there is no reflec-tion, which means R1 is minus infinity.Figure 6 shows the return loss meas-urement result example for an amplifier.

In case of using the S-parameter testset with the instrument, you can per-form the full two-port calibration toobtain the best accuracy for the inputor output impedance and the otherevaluations.

Figure 6. Return Loss Measurement Example

7

Spectrum Evaluation for AmplifierHarmonic Distortion MeasurementNonlinear behaviors in the amplifierwill cause harmonics of the input signal to appear at the output alongwith the fundamental. These harm-onics are integer multiples of theinput (fundamental) frequency, andare usually specified in terms of dBbelow the fundamental signal for agiven input level (commonly expressedas dBc). The measurement configura-tion for the harmonic distortion isshown in Figure 7.

The 4395A/96B has a multiple markerfunction. This function can easilyindicate all of the harmonic distortionpoints. Figure 8 shows the measure-ment example for harmonic distor-tion of an amplifier.

Agilent 4395A/96B

Signal Generator Amplifierunder test

To: Spectrum Port

Figure 7. Measurement Configuration for Harmonic Distortion

Figure 8. Harmonic Distortion Measurement

8

Intermodulation Distortion MeasurementAnother nonlinear behavior in ampli-fiers is intermodulation distortion.When two input signals are applied toa nonlinear amplifier, the output con-tains additional frequency compo-nents. These additional componentsare the intermodulation distortionproducts. As products of a two-toneexample, the output signal will con-tain frequency components at the twofundamental input frequencies [f1 and f2];harmonics will be seen at [2xf1, 2xf2,3xf1, and 3xf2]; second-order prod-ucts at [f1+f2 and f1-f2]; and thirdorder products at [2xf1-f2 and 2xf2-f1].The third order products are veryclose to the fundamentals and typi-cally fall within the bandwidth of theamplifier. To measure intermodula-tion distortion, two signal generatorsare required. The signal generatorsoutputs are connected togetherthrough a coupler or combiner. Thisdevice must ensure sufficient isola-tion between the two RF sources sothat no intermodulation occurs intheir output stages. The measurementconfiguration is shown in Figure 9.

The Agilent 4395A/96B, using the listsweep function, allows you to breakthe span into segments with variablenumbers of points and RBW’s. Thisoptimizes the sweep speed for properdynamic range, and enables you tolook at both the two tone signals and the intermodulation distortions.Figure 10 shows the measurementresults by using the list sweep function.

Agilent 4395A/96B

Signal Generator

Combiner

Signal GeneratorAmplifierunder test

To Spectrum Port

Figure 9. Measurement Configuration of the Intermodulation Distortion

Segment #1 Segment #2 Segment #3

Center Freq.=99.9MHz

Span Freq.=20kHz

RBW=30Hz

Center Freq.=100.05MHz

Span Freq.=200kHz

RBW=1kHz

Center Freq.=100.2MHz

Span Freq.=20kHz

RBW=30Hz

Figure 10. Measurement Results of Intermodulation Distortion by Using List SweepFunction

9

Noise Figure (NF) MeasurementNoise Figure (NF) expresses the de-gree of noise generated in an amplifier.The NF value for an ideal amplifier,with zero internal noise, is 1 (=0dB).The NF value for an amplifier isdetermined with the following formula.

Cin/Nin Cin/kTB NoutNF = = =

Cout/Nout CinG/Nout GKTB

Where, Cin = Input Signal Power [W] Nin = Noise Signal Power [W] Cout = Output Signal Power [W] Nout = Output Noise Power [W] K = Bolzmann’s constant =

1.381 x 10-23 (J/°K) T = Absolute Temperature (°K) B = Noise Bandwidth (Hz) G = Amplifier Gain

The NF value can be displayed on thelog scale by using the following for-mula.

NoutNF [dB] =10 log

GKTB

Nout =10 log -10 log G -10 log (KT)

B

The third item is a constant whichbecomes 10 log (KT) = -174 dBm/Hzwhen the temperature is 23°C. Thus,only the first item (i.e. noise power)and the second item (i.e. an amplifi-er’s gain) have to be measured. Theformula for determining NF can beexpressed as follows.

NF [dB] = Nout [dBm/Hz] - G [dB] +174 [dBm/Hz]

N and G are the noise level andamplifier gain as measured by thecombination analyzers. For example,when Nout = -140 dBm/Hz and G =10 dB, NF is 24 dB. These calcula-tions can be easily customized usingthe built-in IBASIC function.

Filter EvaluationFilters are generally passive, linear,two-port devices that, most often, can be completely characterized usingswept frequency transmission/reflec-tion techniques, and sometimes spec-trum technique.

Figure 11 shows the major measure-ment parameters for filter (especiallyIF SAW filter) evaluation.

IF SAW Filter Evaluation

NACenter FrequencyInsertion Loss3 dB BandwidthRippleGroup Delayetc.

SAIMD

f1 f2f: Center Freq.

F = 2 × f1 − f2

Figure 11. Major Measurement Parameters of IF SAW Filter

10

Network Evaluation for FilterThe measurement configuration fornetwork analysis of a filter is shownin Figure 12. Either an optionalReflection/Transmission test set oran optional S-parameter test set isrequired to perform this analysis.

Insertion Loss/Bandwidth MeasurementFilters are required to pass signalsfalling within a specified bandwidthwith minimum loss and distortion,while rejecting signals outside of thatbandwidth. Insertion Loss is definedin dB as twenty times the log of thetransmitted voltage divided by theincident voltages. (All signals withinthe bandwidth will be attenuatedidentically.) Out-of-band rejection is ameasure of how well a filter passessignals within its bandwidth whilesimultaneously rejecting all other sig-nals outside of that same bandwidth.The Agilent 4395A/96B is suitable for this measurement due to its widedynamic range ( ≥ 115 dB). The band-width search feature automaticallycalculates the center sweep parame-ter value, bandwidth, Q, insertionloss, and cut-off point deviation (fromthe center of a bandpass or bandreject trace). This feature also displaysthe calculation for all of these param-eters. Figure 13 shows the measure-ment example using the bandwidthsearch feature.

S-parameterTest Set

Agilent 4395A/96B Agilent 4395A/96B

Agilent 87512AR/T Test Set

Filter under test

Filter under test

(1) R/T Test Set is used (2) S-Parameter Test is used

Figure 12. Measurement Configuration for Filter Evaluation

Figure 13. Measurement Example by Using the Bandwidth Search Feature

11

Group Delay MeasurementRecently, linear phase performance of a filter’s bandwidth has becomequite important in digital communi-cation. The linear phase performance(linearity of the device’s phase vs. frequency) is evaluated using groupdelay as the measurement parameter.Group delay is a measure of the slopeof the transmission phase response.The formula for the group delay is asfollows:

Group Delay = - delta phase/(360°xdelta freq.)

Where, delta phase: phase difference

between frequency apertures delta freq.: frequency aperture

The frequency aperture (%) must beset for the proper group delay evalua-tion result.

Figure 15 shows the measurementresults of group delay.

Figure 14. Group Delay

Figure 15. Group Delay Measurement Result

Phase Samples 180° Apart

360°

12

Spectrum Evaluation for FilterIntermodulation Distortion MeasurementIntermediate Frequency (IF) SAW fil-ter evaluations require not only thegeneral network filter parameter eval-uation, but also intermodulation dis-tortion evaluation. This measurementis performed by applying two inputsignals to a filter. A third order signalcan then be seen within the filterbandwidth. For example, when thecenter frequency of the filter is f0,and the two input signals are f1 = f0 +100 kHz, f2 = f0 + 200 kHz, the thirdorder signal ((2xf1 - f2) = f0) inputs tothe filter. The size of this third ordersignal can be evaluated after goingthrough the filter using the spectrumanalyzer. The measurement configura-tion for the SAW filter intermodula-tion distortion is shown in Figure 16.

Figure 17 shows the 83.16 MHz SAWfilter intermodulation distortionmeasurement, with two input signals(83.16 MHz + 100 kHz, 83.16 MHz +200 kHz).

As described before, the 4395A/96Bare capable of both the vector net-work analysis and the spectrumanalysis. These abilities contribute to savings in equipment cost, benchspace, and work when performing filter evaluations.

Agilent 4395A/96B

Signal Generator

Signal Generator

Combiner

Filter under test

To SpectrumPort

f = 2 × f1 − f2

f1

f2

f

Figure 16. Measurement Configuration of the SAW Filter Intermodulation Distortion

Figure 17. IMD measurement of SAW filter

13

Capacitor EvaluationIt is important to evaluate electronicdevices at the actual operating condi-tion before performing circuit design.A full-featured impedance measure-ment function can be added to the4395A/96B by adding Option 010 andthe 43961A RF Impedance Test Kit.The 4395A impedance measurementsystem covers from 100 kHz to 500 MHz,while the 4396B impedance measure-ment system covers from 100 kHz to1.8 GHz. The impedance measure-ment system configuration is shownin Figure 18. A variety of 7 mm typeAgilent test fixtures can be used.

Direct Impedance MeasurementThe impedance measurement systemprovides accurate impedance meas-urement with direct impedanceparameter read-out. To achieve accu-rate impedance measurement, thesesystems offer open/short/load andelectrical delay compensation func-tions, which eliminate errors intro-duced by a test fixture. Figure 19shows the impedance measurementresults for a capacitor using theimpedance measurement system.

Agilent 4395A/96Bwith Opt. 010

RF OUT R A B

R A

OUTPUT

RF INDCSOURCEINPUT

N-N CableTest Fixture

Agilent 43961A

Figure 18. Impedance Measurement Configuration

Figure 19. Impedance Measurement Result of Capacitor

14

Equivalent Circuit AnalysisThe impedance measurement func-tion of the Agilent CombinationAnalyzer Family provides equivalentcircuit analysis capability based on athree- or four-element circuit (Figure20). This feature allows Calculation of the impedance parameters for eachequivalent circuit element, and willalso simulate the frequency charac-teristics based on the calculatedparameters.

Figure 21 shows a comparison betweenan actual measurement result and thesimulated result. You can see that thesimulated characteristics are almostthe same as the actual characteristics.

ConclusionThe Agilent Combination Analyzerskeep up with your demands by offer-ing the capabilities you need for yourproduct’s development.

Figure 20. Five Equivalent Circuit Models

Figure 21. Simulation of Frequency Characteristics

Online assistance:

www.agilent.com/find/assist

Phone or Fax

United States:(tel) 1 800 452 4844

Canada:(tel) 1 877 894 4414(fax) (905) 282-6495

China:(tel) 800-810-0189(fax) 1-0800-650-0121

Europe:(tel) (31 20) 547 2323(fax) (31 20) 547 2390

Japan:(tel) (81) 426 56 7832(fax) (81) 426 56 7840

Korea:(tel) (82-2) 2004-5004 (fax) (82-2) 2004-5115

Latin America:(tel) (305) 269 7500(fax) (305) 269 7599

Taiwan:(tel) 080-004-7866 (fax) (886-2) 2545-6723

Other Asia Pacific Countries:(tel) (65) 375-8100 (fax) (65) 836-0252Email: tm_asia@agilent.com

Product specifications and descriptions in this document subject to change without notice.

© Agilent Technologies, Inc. 2001 Printed in USA December 19, 2001 5967-5942E