microwave circuits and design - lakehead...

Engineering 0150

MICROWAVE CIRCUITS and DESIGN

Laboratory Manual

Jason ServaisEl. Eng. Technologist

Department of Electrical EngineeringLakehead University

Thunder Bay, ON

Winter 2014

THINK SAFETY

MICROWAVE CIRCUITS AND DESIGN

Content

Policy and Rules for Laboratory Exercises 4

Exp. #1 Microwave Test Bench 6

Exp. #2 Antenna Polar Plots 12

Exp. #3 Microwave Slotted Line 24

Exp. #4 Introduction to the Vector Network Analyzer 29

Exp. #5 TRL Vector Network Analyzer Calibration 41

References

1. 'Microwave Devices and Circuits'Samuel Y. Liao, Prentice Hall, 2nd ed.

2. 'Microwave Engineering'David M. Pozar, Wiley, 4th ed.

3. 'Experimental Microwaves'A. W. Cross, Sanders Electronics Ltd.

4. 'Microwave Technology'Dennis Roddy, Prentice Hall

5. Agilent AN 1287-9 Application Note: 'In-Fixture Measurements Using Vector Network Analyzers'. 6. Agilent E5071C ENA Network Analyzer Help (online)

3

4Lakehead University

Department of Electrical Engineering

POLICY AND RULES FOR LABORATORY EXERCISES

• No Food, No Beverages allowed in the laboratory room!!• Keep clothes, bags, etc. OFF the benches with equipment on them.• Be On Time ! Being late is annoying to your lab partners and if the experiment has progressed too far

you may not be credited with doing the experiment!• Safety precautions must be observed at all times to prevent electric shock, damage to instruments, etc.

COME PREPARED! BOTH THE WRITTEN LAB REPORTS AND LAB PERFORMANCE (INCL. ATTITUDE, PUNCTUALITY, PREPAREDNESS) WILL BE CONSIDERED FOR THE FINAL LAB MARK!

Lab Exercises

GeneralThe maximum number of students in a lab work group is indicated on the sign-up sheet.Should students leave a work group for whatever reason such that only one student remains in a group, this student may join another team provided there is still room in that team without exceeding the above maximum number. Missed Lab ExercisesIt is mandatory to perform all lab exercises according to course requirements.Failure to perform one or more lab exercises results in a grade of "F" for the course.

When a student misses a lab exercise for whatever reason he/she must notify the instructor as soon as possible. If the reasons given for the absence are satisfactory to the instructor, a make-up opportunity may be arranged. There may be a chance to let the student join another team to perform the missed lab provided this does not then exceed the max number of students in that group. If it is not possible to accommodate this then a final make-up date will be arranged to take place within one week after the end of classes. Should the student fail to attend this appointment he/she will be required to provide sufficient proof of inability to attend (medical certificate, air ticket, etc.) to avoid the "F" grade. The onus of proof lies entirely with the student! The student then must immediately make another appointment with the lab instructor!The lab report in such a case is due within one week after performance, else the "F' stands.

Lab Reports

GeneralIt is mandatory that all lab reports must be submitted on time as specified by the lab instructor! PlagiarismPlagiarism will not be tolerated and may result in a grade of "F" for the course.Any material taken from sources like books, manuals, web sites, magazines, etc. must be clearly referenced as a footnote or under a bibliography!A re-write of a report will be granted only under exceptional circumstances!Missed Lab exercisesIn some courses group lab reports may be allowed by the lab instructor. The group's composition is also determined by the lab instructor.If a student fails to perform a lab exercise with his group he/she will have to write his/her own individual lab report!Late SubmissionThe submission schedule (due date) will be made known by the attending lab instructor.Late submission results in a deduction of 0.5 marks per day out of 10 full marks.No or Partial SubmissionA final date for submission will be clearly indicated on the sign-up sheet and/or announced by the lab instructor. If after that date not all lab reports have been submitted, the student will receive a grade of "F" for the course!

The deduction of marks for late submission still applies.

Format 5• The report must be typed. Graphs may be produced by computer, provided the software is suited for

that use, i.e. grid, proper scaling and correct labelling can be achieved and the plot is smooth. If a graph is used to extract data or to provide some precise information, show precisely how the information is obtained (here it may be better to draw it on graph paper by hand - usually is faster, too).

• For your report use YOUR OWN words to present your report concise, clear and clean.As pointed out above, copying etc. will be considered as plagiarism and will be severely punished by reducing marks or, in severe cases, served with an "F" as mentioned above (the provider/lender of the original work included)!

• The notes/sheets containing the raw data taken by each student during the experiment are to be initialized by the attending technologist before leaving the lab and attached to the written report. Reports with the raw data missing are subject to a deduction of one full mark (= 10%)!

The student is encouraged to develop and use his/her own personal style for writing and presenting his/her report. However, standard procedures in industry and research laboratories require certain information to be documented. Therefore, adhere fairly loosely to a general format like the following:- Title page

(Please make an exception here: Pages stapled - no folders, plastic covers, etc):Course number, Experiment number, Experiment Title, Name, Lab partners, Date of performance

- AbstractStatement of objective of the experiment (one or two sentences)Concise and pertinent outline of the theory underlying the experiment (max one page)

- Experiment and AnalysisIf the experiment consists of two or more parts, keep the experimental and analysis sections together – the reader of your report does not want to continuously flip pages back and forth to look for data etc.!Brief outline of the method of investigation (whatever is applicable): Procedure Measurement techniques Schematic diagramsEquipment identificationData (tables)Observations relevant to the experiment and the resultsArrange experimental data, and do the necessary calculations (if applicable, at least a sample calculation), to prepare for analysisTheoretical calculations (at least sample calculation)Comparison of experimental results with theory (preferably in form of tables/graphs)Probable causes and magnitude of errors

- Conclusion (Summary)- Review questions- Raw data notes, attached to report, and initialized by the attending lab technologist.

GOOD PRESENTATION IS OF THE UTMOST IMPORTANCE, AS IS CORRECT ENGLISH AND GRAMMAR. EXPECT 20% OF YOUR LAB MARK ASSIGNED TO THIS AREA!!

GENERAL INFORMATION

EQUIPMENT

If equipment needs to be signed out, contact one of the technologists. The person signing it out is responsible for it! Your marks will be held back until all equipment, books, data manuals, tools etc. are returned (i.e. your graduation might depend upon it!). Signed-out equipment has to be returned to the same technologist from whom you signed it out! Assure your name is then removed from his sign-out list!No equipment may be removed from any of the laboratories without explicit permission!

As you see, this is very important! TAKE LABS VERY SERIOUSLY!Lakehead University, Electrical Engineering Department July 2007

Microwave Circuits and Design

Experiment No. 1 - 0150 Microwave Test Bench

1. MICROWAVE TEST BENCH

ObjectiveTo gain familiarity and facility with basic microwave components. To find the mode characteristics of a reflex klystron. To determine the frequency and wavelength in a waveguide.

TheoryKLYSTRON

The reflex klystron tube used in this experiment is a complete microwave oscillator in a tube. Klystron tubes where once widely used as local oscillators in receivers. They are still used in high-power, high-frequency applications as they can achieve peak powers around one megawatt. These tubes can oscillate at different frequencies (modes) depending on the DC voltage applied to a reflector plate and can be used a voltage controlled oscillators. The reflex klystron operates in a number of different modes because of the relationship between return transit time of electrons to periods of modulating signal. The relationship between the mode and the repeller (or reflector-) voltage is being given by

Eq.6-3-22, Ref.(1):V 0

(V 0+V r )2=

(2πN−π2 )

2

8ω2L2⋅em

(1)

The operation of the klystron and the generation of klystron modes are shown in Figures 1 & 2.

6



Figure 1.1

Figure 1.2

See Reference [1] chapters 6.2 and 6.3; Reference [2] Section 11.5; Reference [3] Chapter II A, B; Reference [4] Chapter 9.5 for further information.

WAVEGUIDE FREQUENCY AND WAVELENGTH

See Reference [1] Chapter 4.1; Reference [2] Chapter 2; Reference [4] Chapter 4.3.

With respect to the waveguide, the guide wavelength is given by

7



λg=λ

√1−( f cf )2

(2)

where λ is the wavelength in free space, and fc is the cut-off frequency. From the guide dimensions fc may be calculated as follows

TEmn = TE10 mode

f c=1

2√μe⋅√m

2

a2+n2

b2 (3)

f c=cλc (4)

where m is the number of half waves of em energy in the x-direction, and n the number of of half waves of em in the y-direction.

Guide wavelength may be calculated from the VSWR. Figure 5 shows a typical VSWR sketch. It can be seen from this that the average value of distance between minima is half of one guide wavelength.

Standard inner mechanical dimensionsfor X-band waveguides:

a = 2.286 cm (0.9 inches) b = 1.016 cm (0.4 inches)

Figure 1.3

8



Figure 1.4

The wave velocity in the guide is given by

v g=v p

√1−( f cf )2

(5)

where vp is the velocity in free space = 300 x 108 cm/s.

The relation between wavelengths is given by

1

λg2=

1

λ2−

1

λc2 (6)

andλc=2a (7)

Experiment

Equipment

Figure 1.5 shows the basic circuit arrangement for the measurements, using the Marconi Test Bench.

Procedure

CHARACTERISTICS OF A REFLEX KLYSTRON

Use the set-up as shown in Figure 1.5 with the klystron mounted.

The instruments are usually ready to go. If for some reason the set is not ready yet, proceed as follows:Set the Klystron power supply to 'HEATERS'. Wait for an audible click, then the unit is ready for use. Set the receiver to 'MAINS'. For this experiment, switch to 'SQ. WAVE'. That modulates the GHz oscillation with a 1kHz square wave, for which the receiver is tuned to. Fine tune the 'MODULATION FREQUENCY' to achieve a max indication on the receiver's meter.If during the experiment it is desired to stop the radiation switch back to 'HEATERS'.

9



Set the repeller voltage at − 300V and slowly increase the negative voltage by turning anti-clockwise until a peak is observed on the power meter. With the attenuator adjust the meter to an approximately full scale indication. Further increase will produce outputs as shown in Figure 3: Rise, Peak, Fall, then zero until the next bunch occurs.

Measure and plot the power output against the repeller voltage over the full range of repeller voltage adjustment, verifying the characteristics shown in Figure 3.

For the three highest peaks measure the frequency at which the peaks occur. Measure also the frequency at about half point on the rise and about half point on the fall of the output curve. Plot the change of frequency within each power mode against the repeller voltage.

Fig. 1.5

MEASUREMENT OF FREQUENCY AND WAVELENGTH

Using the slotted line section, measure the distances between minima of the VSWR caused by the reflections of the attached antenna. With the reflecting sheet in a fixed position and moving the slotted line probe, measure the guide wavelength. With the slotted line probe in a fixed position and moving the reflecting sheet, measure the free space wavelength.From this find the cut-off wavelength and frequency.

Switch off and disconnect the klystron. Connect the solid state 4 - 8 GHz microwave oscillator. Starting at 8 GHz, tune the oscillator down until the output ceases to exist. Note the frequency and compare with the cut-off frequency and cut-off wavelength, found from the measurements above.

Compare these results with Eq.(3) as well with the physical dimensions of the waveguide.(see wavemeter conversion chart on next page)

10



ConclusionsComment on your experiment.

Fig. 1.6

11


Experiment No. 2 - 0150 Antenna Polar Plots

2. ANTENNA POLAR PLOTS

ObjectiveTo plot the polar diagrams of

- a waveguide horn antenna- a parabolic dish antenna- a dielectric rod antenna

To calculate the gain of a waveguide horn antenna.

TheorySee the manual accompanying the test equipment: "Experiments with Antenna Test Bench Type 6452A Series". Copies of the relevant parts are attached to this manual.See also Reference [2] Section 14.1, and Reference [4] Chapter 10.

PARABOLOID

The main use of paraboloids as antenna is their property of converting a spherical wave arising from a point source at the focus, into a plane wave. This produces a highly directional antenna with a narrow beam-width. As with horn antennas some spillover occurs as shown in Figure 2.1.

Figure 2.1

The waveguide feed to the paraboloid used in this experiment is a dipole. The radiation from the waveguide drives two dipoles as shown, A being the driven element and B, the reflector. Ideally, the dipole and reflector are equivalent to two half wave dipoles spaced by 1/4 wavelength.

12



WAVEGUIDE HORN

The waveguide horn directs the radiation into a narrow lobe. The width of this lobe is a function of the horn dimensions. Certain assumptions may be made regarding the phase differences at the horn aperture and the beam-width of the horn can be approximately stated as

E-phase beam-width = 0.88 λ/aH-phase beam-width = 0.88 λ/b

where a and b are the dimensions of the aperture.

It will be noted that small variations appear across the aperture and when the polar plots are drawn it will be seen that "spillover" effects occur. Radiation caused by horn currents creep around the edges of the horn.

DIELECTRIC ROD

The dielectric rod used in this experiment is a solid rod of dielectric material excited by an open ended waveguide as shown in Figure 2.2.

Figure 2.2

All dielectric antennas have the common property that they produce single lobe radiation patterns directed along the axis of the antennas and the directivity is proportional to their length. Also dielectric antennas when used in arrays have the important property that their radiation pattern changes little with frequency - they are broadband devices.

ANTENNA GAIN

The general theory of the measurement of the gain of a waveguide horn is given in the attached copy of the test bench manual.

13



Experiment

Equipment

Sanders Antenna Test bench Type 6452A

Procedure

POLAR PLOTS

Figure 2.3

Set up the system as shown in Figure 2.3, starting with the paraboloid antenna.

Initially set up the system as described in the Procedure for Experiment no.1.Then move the receiving antenna such that the receiver indicates max received power. If necessary, correct the position indicator to exactly 1800.Turn the calibrated attenuator to position 3.5.Now adjust the receiver such that the pointer sits right in the middle, at '50' (top scale).

On the attenuator's Calibration Sheet, look up the dB attenuation inserted at this attenuator position. This is your reference for the received signal strength.

Move the receiving antenna for 2 degrees: The receiver indicates a loss of signal strength. Replace this loss by taking out attenuation until the receiver indicates exactly the same signal strength as before, i.e. the pointer is again in the middle. Read the new position of the calibrated attenuator and on the chart find the corresponding attenuation in dB now inserted into the line. The difference in dB to the reference represents the loss of signal strength at this angle. Plot this onto the polar diagram.

Repeat this procedure every 2 degrees for 30 degrees. The resulting plot on the polar diagram shows the radiation pattern for this particular antenna.

14



In the same manner, perform measurements with the other two antennas:for the waveguide horn: every 50 for 600

the dielectric rod: every 50 for 600

GAIN OF WAVEGUIDE HORN

Measure the klystron frequency/guide wavelength using the wavemeter. To find the wavelength required for the calculations (free space wavelength!), the same conversion sheet as in Experiment 1 may be used. Measure the aperture dimensions of the horn antenna and the distance between the apertures.

Measure the gain of the waveguide horn antenna according to the "Experimental Procedure" presented in the attached excerpt from the Sanders Manual.

Analysis1. From the plots find the 3 dB-beamwidth of each antenna. Comment on the

performance of the antennas.

2. Calculate the theoretical gain for the horn antenna as outlined in the excerpt from the Sanders Manual "Theoretical Gain of a Horn".For the antennas used:

a = b = 7.62/λ0 a, b, λ0, all in cm lE = lH = 16.56/λ0 lE, lH, λ0, all in cm

To find λ, use eqn.(6) from Exp.1, with λc = 4.57cm

Compare with the gain found from your measurements, as outlined in the excerpt from the Sanders Manual "Experimental Procedure", including the corrections.

3. The free-space attenuation is given by:

−20log10(4πSλ ) dB (1)

From your measurement, the free-space attenuation is:

PT+GT+GR−PR dB(2)

where PR = power received Eqs. (1) and (2) taken from Ref. (1)

PT = power transmittedGT = gain of transmitting antennaGR = gain of receiving antennaS = distance between apertures of the antennas

Also, find the free-space attenuation from the nomogram from Ref. [2] on page 23.

15



(Extend the logarithmic meter-scale as good as you can since the actual distance between antennas in our experiment is < 3 m)

Compare thee three results.


16



Fig. 2.4

17



18



19



20



21



22


Experiment No. 3 - 0150 Microwave Slotted Line

3. MICROWAVE SLOTTED LINE

ObjectiveTo

a) find the wavelength λ using the slotted lineb) find the impedance of an unknown loadc) match the load by means of a double-stub tunerd) become familiar with the use of the Smith Chart

TheorySee Reference [1] Sections 3.5 and 3.6; Reference [2] Sections 2.4 and 5.3; Reference [4] Chapter 3.

The maximum amplitude of a wave being propagated along a two-conductor transmission line can be shown to be

V max=V + e−αz

+V−eαz (1)

and the minimum amplitude

V min=V +e−αz

−V−eαz (2)

V max occurs at βz=nπ [n=0, ±1, ±2, . . .]

V min occurs at βz= (2n−1 )π2 [n=0, ±1, ±2, . . .]

with α = attenuation coefficientβ = phase shift coefficientz = distance along transmission line

The distance between any two successive maxima or minima is one-half wavelength.

STANDING WAVE RATIO

This is defined to be

VSWR=∣Vmax

Vmin

∣ (3)

23



The reflected wave is defined to be the result of an incident wave and its reflection coefficient:

VSWR=1+ρL

1−ρL(4)

and

ρ=VSWR−1VSWR+1 (5)

Hence the reflection coefficient can be determined from the VSWR by means of the Smith Chart.

GENERAL RULES FOR STANDING WAVE PATTERNS (see Figure 3.1)

Figure 3.1

1. The shift in the minimum after the load has been shorted is never more than ±λ/4.

2. If shorting the load causes the minimum to move towards the load, a capacitive component exists in the load.

3. If shorting the load causes the minimum to shift towards the generator, an inductive component exists in the load.

4. If shorting the load causes no shift in the minimum, a completely resistive load exists equal to Z0/VSWR.

5. If shorting the load causes the minimum to shift exactly ±λ/4, the load is completely resistive and has a value of Z0/VSWR.

6. When the load is shorted, the minimum will always be a multiple of λ/2 from the load.

24



The Smith Chart

This is a graphical method for solving transmission line problems. It is a plot of the normalized impedance and admittance with the angle and magnitude of a generalized complex reflection coefficient in a unit circle.

The reflection coefficient for a transmission line at the load end is given by

Γ L=Z L−Z0

Z L+Z0

=ρLeiθL

(6)

ZL = load impedanceZ0 = characteristic impedanceΦL = the phase angle between incident and reflected wave

The chart is organized with the reflection coefficient as the radial coordinate and the circles concentric with the centre of the unit circle are circles of a constant reflective coefficient. These circles are also the contour of constant VSWR. Since the VSWR is never less than unity the scale for the VSWR varies from one to infinity on the real axis. Distances are given in wavelengths towards the generator and also towards the load.

SMITH CHART CHARACTERISTICS

1. The constant r and constant reactance x loci for two families of orthogonal circlesin the chart.

2. The constant r and constant reactance x circles all pass through the point (Γr = 1,Γi = 0).

3. Upper half represents +jx.

4. Lower half represents -jx.

5. For admittance, the constant r circles become constant g circles and the constant x circles become constant susceptance b circles.

6. The distance around the chart once is one-half wavelength (λ/2).

7. At a point of Zmin = 1/VSWR, there is a Vmin.

8. At a point of Zmax = VSWR, there is a Vmax.

9. The horizontal radius to the right of the chart centre corresponds to Vmax, Imin, Zmin and VSWR.

25



Note:Zmax=

Zmax

Z0

; Zmin=Zmin

Z0

10. The horizontal radius to the left of the chart centre corresponds to Vmin, Imax, Zmin and 1/VSWR.

11. Since the normalized admittance Y is a reciprocal of the normalized impedance Z, the corresponding quantities in the admittance chart are 1800 out of phase with those in the impedance chart.

12. The normalized impedance or admittance is repeated for every half-wavelength of distance.

13. We know that

ρ=VSWR−1VSWR+1

14. A "slotted line" can be used to measure a standing wave pattern directly and the magnitudes of the reflection coefficient, reflected power, transmitter power and load impedance may be calculated using the Smith Chart.

Experiment

Equipment

Slotted Line and attachments, general Radio (see Figure 3.2)

Figure 3.2

ProcedureMake sure the Generating Oscillator (the left one) is set to ~500 MHz. Adjust the Local Oscillator (the right one) to ~470 MHz and fine tune such that the IF Amplifier is adjusted to exactly 30 MHz, i.e. to a max. indication [Heterodyne principle, with mixer].

1. Note the load number!

26



2. Determine the frequency and wavelength by measuring the distance between minima using the short circuit load.

3. Determine the impedance of the unknown load by means of VSWR measurement, minimum shift when shorted, and by use of the Smith Chart.

4. set up the double-stub tuner as shown in Figure 3.

5. Using the double-stub tuner and the Smith Chart obtain and show a match for the load with the transmission line.

6. Compared to the original mismatch and VSWR when the load was connected directly to the line the VSWR should have dramatically improved after completion of Step 4. Try now to manipulate the studs carefully to obtain the final optimum match.

7. State the final VSWR achieved.

Figure 3.3

AnalysisShow the Smith Chart work and the necessary calculations to achieve the match.


.

27


Experiment No. 4 - 0150 Introduction to the VNA

4. INTRODUCTION TO THE VECTOR NETWORK ANALYZER

ObjectiveTo gain familiarity with a Vector Network Analyzer (VNA). To perform a 2-port calibration procedure. To measure the reflection coefficient of one-port networks and the S-parameters of two-port networks.

Theory

See Reference [2] Section 4.3 and Reference [4] Chapter 2, or other literature dealing with RF, transmission lines and microwaves, for further information.

An unknown linear two-port device can be described by a set of parameters computed from measurements. For low-frequency characterization of devices, the three most commonly measured parameters are the H, Y and Z-parameters. This requires measuring voltages and currents as a function of frequency at the input or output nodes (ports) of the device, and the use of open circuits and short circuits as part of the measurement. At high frequencies, however, it is very hard to measure total voltage and current at the device ports. That is why S-parameters were developed which relate to measurements such as gain, loss and reflection coefficient. They are defined in terms of traveling waves at the instrument ports, which are relatively easy to measure, and don’t require connection of open or short loads. If desired, H, Y, or Z-parameters can be derived from S-parameters.

An n-port device has n2 S-parameters. Thus, a 2-port device has four S-parameters.The numbering convention is that the first number following the “S” is the port where the signal emerges, and the second number is the port where the signal is applied. For example, S21 is a measure of the signal coming out of port 2 relative to the RF stimulus entering port 1. When the numbers are the same (e.g., S11), it indicates a reflection measurement, as the input and output ports are the same.

28



Fig. 5.1

Also

Fig. 5.2

Formula set :

S11 = forward reflection coefficient (input match)

29



S22 = reverse reflection coefficient (output match)S21 = forward transmission coefficient (gain or loss)S12 = reverse transmission coefficient (isolation)

S-parameters are inherently linear quantities – however, they are often expressed in a log-magnitude format.

S11 and S22 are often displayed on a Smith Chart.

S11 and S21 are determined by measuring the magnitude and phase of the incident, reflected and transmitted voltage signals when the output is a perfect Z0 (a load that equals the characteristic impedance of the test system). This condition guarantees that a2 is zero, since there is no reflection from an ideal load. S11 is equivalent to the input complex reflection coefficient or impedance of the DUT (device under test), and S21 is the forward complex transmission coefficient. Likewise, by placing the source at port 2 and terminating port 1 in a perfect load (making a1 zero), S22 and S12 measurements can be made. S22 is equivalent to the output complex reflection coefficient or output impedance of the DUT, and S12 is the reverse complex transmission coefficient.

The accuracy of S-parameter measurements depends greatly on how good a termination is applied to the port not being excited. Anything other than a perfect load (Z0

= 50Ω !) will result in a1 or a2 not being zero (which violates the definition for S-parameters). For this reason, error correction is very important for accurate S-parameter measurements.

Experiment

Equipment

RF Network Analyzer Agilent 8712ESCalibration Kit H&P 85032ECircuit elements, coaxial cable, and connectorsGrounding cable with wrist band

Procedure

• Wear the wrist strap to avoid electrostatic discharge at all times, especially when doing connections.

• Handle connectors with care. Do not drop, hit or turn connectors (turn only the outer side when connecting).

• Try not to bend cables too much.

Part A ONE-PORT MEASUREMENT AND VNA CALIBRATION

For accurate measurements, the VNA must be calibrated to remove the effect of cables and connectors from the instrument to the device under test (DUT). To illustrate this, we'll compare the measurements of a short circuit reference without and with calibration. Calibration methods can be classified into 1-Port and 2-Port. The first is simpler and

30



faster to perform, but it is only useful for 1-Port measurements (impedance or reflection coefficient measurements). The calibration kit contains three reference loads: Open circuit, Short circuit, and a 50Ω matched termination, all with male Type-N connectors. Treat these loads very carefully.

1. Press [Preset] to begin. This reverts the VNA settings to a default state.2. Connect the Agilent Tech. 50136 Cable to Port 1. This cable has a male and

female Type-N connectors.3. Set frequency range to [50, 900] MHz.4. Connect short circuit load to the end of the cable as shown in Figure 5.3.5. To use the internal calibration, press [Cal], [Default 1-Port]. This calibration was

performed using full band and 401 frequency points. It corrects systematic errors due to frequency response, load match, source match, and directivity. The calibration reference plane is Port 1 (note the cable is not included in this calibration).

6. Press [Meas 1], [S11], [Format], [Smith Chart]. Store the measurement in memory for later use: [Display], [Data->Memory].

7. Perform a 2-port calibration including the cable: [Cal], [User 2-port], [User 2-port]. Follow the instructions on the screen. This calibration procedure is often referred as SOLT calibration (short, open, load, through). Note: a male-male adapter is needed when asked to connect the “through” load. Since this adapter has a small electrical length, it will introduce a small phase error in the calibration.

8. Re-connect and measure the short circuit reference at the end of the cable connected to Port 1. Press [Display], [Data and Memory] and compare the two measurements. Save this screen (use [Display] [Inverse Colors] for a white background).

9. Do not remove the cable and keep the 2-port calibration settings for the remainder of the experiment.

Fig. 5.3

PART B TRANSMISSION LINE MEASUREMENT

1. Connect the A. T. 50253 Cable between the already cable-extended Port 1 and Port 2.

2. Press [Meas 1], [S21], [Format], [Mag dB], [Scale], [Autoscale]

31



3. Press [Meas 2], [S21], [Format], [Phase], [Scale], [Autoscale]4. Save this screen. Use [Markers] for the following items.5. Measure the cable attenuation at 100 and 800 MHz.6. Measure the change in frequency required to produce a 180 degree phase shift.

PART C FILTER-ATTENUATOR

1. Connect the provided filter (note the part number printed on it). 2. Measure S11 [Meas 1] and S21 [Meas 2]. Use the ‘Mag Log’ formats.3. Adjust [Scale], [Frequency] [Start] [Stop], to optimize the display. Remember that

the frequency range should not be set outside of the calibrated range of [50, 900] MHz. The forward transmission coefficient shows the familiar shape of the filter. Use the [Marker] functions to measure the S-parameters outside and inside the pass band, as well as center frequency and bandwidth.

4. Find the approximate center frequency and the bandwidth (-3dB points) if the filter is a band pass type, or the cut-off frequency for a LP or HP filter.

5. Save the two traces in memory: [Display], [Data->Mem], [Data and Memory] for each.

6. Insert the attenuator on the Port 1 side. Save the display.

Analysis

Part A:

Explain the observed differences in the S11 measurement of the Short in magnitude and phase. Compare the results with an ideal short circuit and explain any mismatch.

Part B:

The cable physical length is 50 cm. Use the measured S21 magnitude value to estimate the attenuation constant in the cable as a function of frequency. Use the measured S21 phase to estimate the phase constant in the cable as a function of frequency. Are the results consistent with theory?

QuestionsCan you trust the measurements of a VNA if you do not know how it has been calibrated? Explain.

Describe in words the meaning of each S-parameter measured in Part C. Compare your measurements with S11 and S21 of an ideal filter of the type you measured.

Explain the observed effect of the attenuator on S11 and S21, both inside and outside the filter band.

Can you measure S11 and S22 on a two-port device simultaneously with this VNA? Explain what actually happens if this combination is set in the instrument.

32



Conclusions

Comment on your experiment.

33


Experiment No. 5 - 0150 TRL Calibration

5. TRL VECTOR NETWORK ANALYZER CALIBRATION

ObjectiveTo gain familiarity with planar microwave circuits and non-coaxial measurements with a VNA. To perform a TRL calibration procedure to remove the effects of cables and connectors. To use measured parameters in a circuit simulator. To compare real and simulated CPW components.

Theory

See the theory section of Experiment 4 for a background on scattering parameters and the vector network analyzer.

See Reference [2] Example 4.7, Reference [5] and other literature dealing with RF, making measurements of devices without standard coaxial connectors for further information. For information about operating the E5071C VNA read the instrument help file online from the Agilent website or press the [Help] button on the instrument.

Making good measurements on devices without standard coaxial connectors is difficult because commercial calibration kits and standard error-correction routines found in most network analyzers can not be used to remove the effect of of the transition between connectors and the device under test (DUT).

In this experiment we will perform a thru-reflect-line (TRL) calibration procedure to remove the effects of cables, connectors and discontinuities when measuring devices mounted on a PCB fixture. TRL calibration is more convenient than SOLT at frequencies higher than 1 GHz. This is because it is difficult to implement the short, open and load references over a wide bandwidth. For example, a microstrip open circuit behaves like a capacitive load. On the other hand, TRL calibration requires just one reference: a transmission line with a known characteristic impedance. Example 4.7 in Reference [2] explains the theory behind the TRL calibration. The E5071C VNA has the calibration equations built-in and after the calibration procedure is performed it automatically de-embeds the error boxes up to the reference plane. The reference plane in the provided PCB board is located at 2.5'' from the SMA connector plane (coincident with the end of the line for the open and short-circuit loads).

34



Experiment

Equipment

Agilent E5071C Vector Network AnalyzerPCB fixture with TRL calibration standards and other loads2 Coaxial cables with Type-N/SMA connectorsGrounding cable with wrist band

Procedure

• Wear the wrist strap to avoid electrostatic discharge at all times, especially when doing connections.

• Handle connectors with care, specially the small SMA connectors in the PCB. Do not drop, hit or turn connectors (turn only the outer side when connecting).

• Try not to bend cables too much.

• Treat the PCB board carefully

1. The VNA accepts input from the front panel buttons, the touch screen and also from the keyboard and mouse.

2. Press [Preset] to begin. This reverts the VNA settings to a default state.

3. Set frequency range to [0.6 GHz, 4 GHz]. In the “Stimulus” section of the instrument press [Start] 600M, [Stop] 4G.

4. Select [Calibration]

5. Select calibration kit definition for the PCB board: [Cal Kit], [Experiment 5]

6. Perform calibration: [Calibrate] [2-port TRL calibration]

1. Measure “thru” standard: connect thru between Ports 1 and 2, select [Thru/Line] [1-2 Thru/Line]

2. Measure “reflect” standard (use the same in Port 1,2): connect open to Port 1 and select [Port 1/Reflect] [Reflect-open], connect the same open to Port 2 and select [Port2/Reflect] [Reflect-open].

3. Measure “line” standard: connect the line between Ports 1 and 2, select [1-2 Line/match]

4. Press [Done] to activate the calibration corrections.

2. Leave the transmission line standard connected. We will save the 2-port S parameters of the line in Touchstone format. First measure [S11] and [S21] in linear polar format: press [Measure], etc... Make sure the measurements are consistent with what you would expect for a

35



section of a matched low-loss transmission line. Then press [Save/Recall], [Save SnP] [S2P]

3. Measure S11 for the short-circuit (use linear Smith Chart format). Save the measurement in Touchstone format: [Save/Recall], [Save SnP], [S1P].

Analysis

Compare the measured transmission line and short-circuit parameters with the parameters predicted by the CPW models in the QUCS simulator. The figure below shows a sample schematic for the short-circuit (some parameters may be different). Compare the magnitude and phase of S11 (measured short-circuit) and S22 (simulated short-circuit) using two Cartesian diagrams, one for magnitude and one for phase.

The substrate and transmission line parameters are: er=3.55, H=0.81 mm, T=35 um, sigma = 5.96e7 S/m2, tand = 0.0021, W = 1.727 mm, S = 1.219 mm, L = 16.84 mm

Perform a similar comparison for the line. Compare magnitudes of S11 and S21 and also phase of S21.

Questions

What are the underlying assumptions used in TRL calibration?

36



What additional measurements could you make to find out if the TRL calibration standards are accurate?Are the measurements consistent with the simulation models? List and explain possible causes may explain discrepancies.

Conclusions

Comment on your experiment.

37

microwave circuits and design - lakehead...

Documents