August 1998 35

Among the many measurementtools sought by the amateurexperimenter, the most de-sired—but generally considered

the least accessible—is the radio-frequencyspectrum analyzer or SA. This need not be.Simple and easily duplicated, this home-built analyzer is capable of useful measure-

By Wes Hayward, W7ZOI, and Terry White, K7TAU

1Notes appear on page 43.

A Spectrum Analyzer forthe Radio AmateurGood tools are priceless when youneed them. Here’s a piece of testequipment you’ve always wantedfor your workbench. Now you canhave it—without spending a fortune.

ments in the 50 kHz to 70 MHz region. Thedesign can be extended easily into the VHFand UHF region with methods outlinedlater. The instrument is configured to beself-calibrating, or capable of calibrationwith simple home-built test gear.1

We often read and hear about “simpledesigns.” Simplicity implies that some-thing is eliminated to make the equipmenteasier to build, use or afford. Unlike de-signs that sacrifice performance for costand simplicity, this one sacrifices only con-venience, while retaining the capabilitiesneeded for accurate measurements.

Figure 1—Block diagram of the spectrum analyzer. The circuit is a double-conversion superheterodyne design with intermediatefrequencies of 110 and 10 MHz.

Modern technology eases the construc-tion of this spectrum analyzer. The loga-rithmic amplifier uses an IF amplifier ICfound in cellular telephones and includes areceived signal strength indicator (RSSI)

Figure 2—Time base for the spectrum analyzer. Refer to the text for a discussion of the various circuit functions. Front-panel controlsinclude SWEEP RATE, SPAN and TUNE. Unless otherwise specified, resistors are 1/4 W, 5% tolerance carbon-composition or film units.Equivalent parts can be substituted.U401, U402, U403—LM358 op ampD403, D406—6.2 V Zener diodes, 1 WC401—Metal film or Mylar, 1.0 µF

capacitor

R420, R423—PC-mount trim pots, 5 kΩ or10 kΩ suitable

R3, R4, R5, R6—Panel-mounted linear

control, 5 kΩ or 10 kΩ suitable. If a 10turn pot is used for R3, R4 is notneeded.

function. Hybrid and monolithic IC build-ing blocks are employed extensively. Theseinclude mixers, amplifiers and VCOs—allvital elements in an analyzer. Finally, it isa rare devoted experimenter today who

does not own an oscilloscope. With goodbasic ’scopes available for about the priceof a hand-held FM transceiver, every ex-perimenter should have one. Our spectrumanalyzer uses a ’scope as the display. There

36 August 1998

are no special requirements for ’scope per-formance other than an X-Y mode with dccoupling in the X and Y axes.

Some Spectrum-Analysis BasicsThe RF spectrum analyzer is essentially

a swept receiver with a visual display. Thedisplay shows the strength of all signalswithin a user-defined frequency span. Eachsignal is represented by a line or blip thatrises out of a background noise, much likethe action of an S meter. Commercial ana-lyzers are calibrated for signal power, withall signals referred to a reference levelat the top of the screen. Our analyzer isdesigned for a basic reference level of–30 dBm, a common value in commercialanalyzers.2

Signal levels are read from the displayby noting that power drops by 10 dB foreach major division on the ’scope. You canchange the reference level. Adding gain tothe analyzer moves the reference to lowerlevels; introducing attenuation aheadof the instrument moves the reference tohigher power levels.

Circuit OverviewFigure 1 is a block diagram of our spec-

trum analyzer. A double-conversion super-heterodyne, it begins with a step attenua-tor, followed by a low-pass filter and thefirst mixer, where incoming signals areupconverted to a 110 MHz first IF. Aftersome gain and band-pass filtering, a sec-ond conversion moves the signals to a10 MHz IF. The resolution bandwidthsavailable are 30 kHz and 300 kHz. A videofilter smooths or averages noise. The avail-able frequency spans range from a per-division maximum of 7 MHz to about50 kHz. The center frequency can be ad-justed over the entire 70 MHz range. Anuncalibrated SPAN control allows expan-sion of the display about the screen center.An uncalibrated SWEEP RATE controlallows the sweep to be controlled andmatched to a given span while avoidingexcessively fast scans that could introduceerrors.

Ideally, a receiver’s first IF should begreater than twice the highest input fre-quency, a design rule that we bend in thisapplication. The input tuning range in-cludes all HF amateur bands and 6 meters.(We’ll discuss higher tuning ranges later.)We picked the 10 MHz second IF becausesurplus-crystal filters and LC filters for thisfrequency are easily built. You can easilyadapt the design’s IF to 10.7 MHz, or otherclose, convenient values.

The swept LO tunes from 110 to180 MHz with a commercial VCO module.The VCO output is amplified to drive ahigh-level-input mixer. The commercialVCO is a recent modification to a designthat started with a homebrew oscillator.3

Amplifiers are included at the 10 and110 MHz IFs. These establish signal levelsthat properly match the log-amplifier win-

Figure 3—An experimental logarithmic amplifier breadboarded to evaluate performanceprior to analyzer construction. You may want to duplicate this circuit and analyze itsperformance if you decide to use other log-amp ICs.

dow, while preserving system dynamicrange. The proper distribution of gain, se-lectivity and signal-handling capability (in-tercepts) of the amplifiers and mixers isvital to achieving good performance in aspectrum analyzer, and indeed, any re-ceiver. A proper design will have the samenumber of stages as a poor one, but willprobably use different components andconsume more current.

The analyzer uses a ±15 V power supply.The positive supply delivers about 0.5 A.The negative supply current drain is under50 mA.

Following sections present the circuitblocks in greater detail, in the order thatthey should be built. The partial but grow-ing system can then be used to test the othersections as they are built, turned on andintegrated. We strongly discourage build-ing the entire analyzer before testing spe-cific sections. Such an approach may workfor casual kits, but is not suitable whencareful control of signal levels is required.That approach also robs you (the builder)of the excitement of the process: the learn-ing that comes from detailed examination.

Before jumping into the circuit details,we reemphasize that this analyzer—al-though simple—is intended for seriousmeasurements. This means that a normalmaximum span display contains no spuri-ous signals. When clean (well-filtered, har-monic-free) signals are applied to the ana-lyzer, there should be no extra products aslong as the signal level is kept on screen.This performance goal applies for a singletone, or for two equal signals at the top ofthe screen.

Time BaseFigure 2 shows the analyzer time base,

designed for basic functionality withoutfrills; the result is a circuit using only ahandful of op amps.4 U401A and U401Bform a free-running sawtooth generator, a

circuit commonly found in function gen-erators. U401A operates as an integrator;current is pulled from the inverting inputthrough a 56-kΩ resistor connected to theSWEEP RATE control. This current mustflow through the capacitor (C401), creat-ing a linearly changing op-amp output volt-age. This ramp is applied to U401B, a re-generative comparator, which provides areset signal to the integrator. The sawtoothwaveform (pin 1 of U401A) is asymmetri-cal: The positive-going ramp grows with aslope determined by the front-panel-mounted SWEEP RATE pot, while the nega-tive-going, faster reset ramp is determinedby fixed-value components.

The U401 ramp is used twice. U402A andB process the ramp to generate a signal thatdrives the ’scope’s X axis. The signal has a0 V-centered range with just over a 10 Vtotal swing. Some of the “square wave” fromthe basic time base (U401B, pin 7) is addedto the input of U402B to cause the sweep toreset quickly, even though the return sweepfor the VCO occurs in a more stable, smoothway. A slight overscan is generated for the Xaxis, serving to hide an aberration occurringnear the sweep beginning.

The sweep also generates the signalthat controls the VCO. The sweep signal(U401A pin 1) is applied to a SPAN control.When the analyzer is set for maximumspan, the VCO voltage (about 2 to 10 V)generates a sweep from 110 to 180 MHz.The VCO uses only positive sweep volt-ages, so the output of U403B is diode-clamped to prevent negative output. Thecenter frequency TUNE, FINE TUNE and aMAX SPAN calibration pot set up the propersweep for maximum span. As the span isreduced with the SPAN control, the sweepexpands on (or zeroes in on) whatever ap-pears at the center of the screen, determinedby the tuning. The center frequency mustbe set for 35 MHz at maximum span, whichcoincides with having the zero signal, or

August 1998 37

“zero spur” at the left edge of the screen.Setting up the time-base function is gen-

erally straightforward. The ’scope can beused to debug, check and study the circuits.The X-axis signal is a ramp ranging from–6 to +6 V with a reset to –15 V during theretrace. A similar ramp appears (without areset pulse) at the VCO output, but with anamplitude dependent on the SPAN controlsetting.

Although the op amps are carefully by-passed, and the signal that tunes the VCO isshielded, most circuits are noncritical.Normal op-amp circuit precautions aretaken with resistors injecting signals intoinverting inputs positioned close to theop-amps.5

A 10-turn front-panel-mounted pot isused for the TUNE control (any value from5 kΩ to 50 kΩ is suitable). A single-turnpot can be substituted if a 10-turn pot is notavailable. A fine-tuning function is in-cluded in this design, but may be omitted ifa 10-turn pot is used for the main tuning.

Log Amplifier and DetectorCentral to any spectrum analyzer is a

logarithmic amplifier. The need for loga-rithmic processing becomes clear if weconsider the range of signals we wantto measure: At the low end, we may wantto look at submicrovolt levels: under–107 dBm in a 50-Ω system. At the otherextreme, we may want to measure the out-put of small transmitters, perhaps up to apower of 1 W, or +30 dBm. The differencebetween the two levels is 137 dB. The hu-man ear is capable of handling linearranges well over 60 dB.6 This is a widedynamic range world and linear displays,such as our screen, are inadequate unlesssome form of data compression or loga-

Figure 4—Transfer characteristics for three different logarithmic amplifier ICs. Althoughthe MC3356 is used in our analyzers, use of the AD8307, shown in the lower curve, isrecommended. Some curves have been linearly scaled to ease comparison.

rithmic processing is used.The circuit element we use for this pro-

cessing is the log amplifier.7 The term is amisnomer, for the usual log-amp IC is botha logarithmic processor (amplifier) and adetector. The chips provide a dc outputvoltage that increases in proportion to thelogarithm of the input amplitude. The cen-tral sensitivity specification for a log ampis a voltage slope that is equal to the volt-age change (per decade or per decibel) ofinput-voltage-amplitude change.

An experimental log amp is shown inFigure 3. We breadboarded and tested thiscircuit to evaluate the log IC. To producethe MC3356 curve shown in Figure 4, the10 MHz output of an HP-8654 signal gen-erator was applied through HP-355 stepattenuators. Exact dc output levels are in-significant, for they can be adjusted withdc voltage gain in a following amplifier.The salient detail that we observe is the dy-namic input window. The MC3356, with a50-Ω input termination, produces a nearlystraight-line output voltage versus inputpower for inputs in the –80 to –10 dBmrange. Hence, the analyzer log amp shouldoperate with an input signal of –10 dBm forsignals at the top of the screen.

We evaluated two other ICs. One, thecommonly available NE/SA604, showsconsiderable ripple. The best performanceoffered came from a recently introducedchip from Analog Devices: the AD8307.This IC is designed specifically for mea-surement applications and offers outstand-ing logarithmic accuracy, a dynamic rangeexceeding 90 dB and better temperaturestability than found with the usual cellular-receiver chips. The AD8307 requires a highdrive level, so it must be preceded withhigher-power amplifiers or impedance-

transforming networks. The bandwidth ofthe AD8307 is about 500 MHz, so care isrequired in its use.

Our analyzer uses the inexpensive andreadily available MC3356 log amp shownin Figure 5.8 An op amp, U303, used to in-crease the signal output to 0.5 V per divi-sion, follows the log chip, U301. The 0 Vlevel corresponds to the bottom of thescreen; a signal of 4 V brings the responseto the top of the screen. The op-amp outputis slightly higher than this, but is then at-tenuated with a LOG AMP CAL control, R2.This pot should be accessible from the out-side of the instrument.

The log amp is preceded by an IF ampli-fier, Q301 through Q303. These stages arebiased for relatively high-current operationto preserve linearity. Gain is controlledthrough variable emitter degeneration inthe form of a PIN diode, D301. Most com-mon 1N4000-series power rectifiers workwell for gain control. The IF GAIN ADJ con-trol (R1) should be available from the exte-rior of the RF-tight amplifier box. We haveplaced it on the front panel of our analyzers.

Calibration of the IF and log amplifier isstraightforward. First, set the ’scope’s Yaxis to 0.5 V/division and short it. Set thenow-working time base to drive the X axisand adjust the ’scope’s vertical positioncontrol to place the horizontal line at thebottom of the screen. Inject a −10 dBm sig-nal from a signal generator into the logamplifier input, remove the short circuitand adjust R2, LOG AMP CAL , for a full-screen (reference level) response. The in-put level is next reduced in 10 dB steps.The horizontal sweep line should dropdown one major division for each 10 dBreduction over a 60 dB range. If this doesnot happen, repeat the procedure with aslightly different drive level. In our analyz-ers, a typical drive level of –13 dBm pro-duced good accuracy.

Now, attach the IF amplifier to the logamplifier and drive them with an inputlevel of –23 dBm. Peak the IF output filterfor maximum response and set R1, IF GAINADJ , for a full-screen response. A true fil-ter peak can be confirmed by varying thegenerator frequency. There is considerableextra range in the IF GAIN ADJ , providingextra flexibility during use.

Resolution FiltersContinuing the backward progression

through the system, we encounter theresolution-bandwidth-determining filters.Our analyzer uses bandwidths of 30 and300 kHz, provided by crystal and LC fil-ters, respectively. The 300 kHz LC filter,the crystal filter and the relay circuitry forbandwidth switching are shown in Figure6. Although shown as individual modules,they can be incorporated into one. The PCboard for the filter includes the LC filterand switching relays with room for a user-selected crystal filter. Builders may wantto implement their own scheme here. We

38 August 1998

C309—Plastic dielectric trim cap (Sprague-Goodman GYD65000)C307, C308, C310—Silver mica or NP0 ceramic capacitors, 10% toleranceC316—0.22 µF ceramic

Figure 5—The 10 MHz IF amplifier and log amplifier used in the analyzer. Refer to the text for adjustment details. Unless otherwisespecified, resistors are 1/4 W, 5% tolerance carbon-composition or film units. Equivalent parts can be substituted.

D301—PIN diode; 1N4007 usedL301—1.35 uH, 18 turns #24 enameled wire on T-44-6 core, Q >150Q301, Q302, Q303—2N3904

R1—Panel-mount, 1 kΩ linearR2—Panel-mount, 5 kΩ linearU301—Motorola MC3356U302—78L05 +5 V regulatorU303—CA3140 op amp

reasoned that builders would want to imple-ment their own ideas. Maintain reasonableshielding for this part of the system. Addi-tional attenuator pads can be inserted in linewith one filter or the other to approximatelyequalize filter loss in the two paths.

You may want to build crystal filters foryour analyzer.9 The VCO stability in thisanalyzer will support resolution band-widths as narrow as 3 to 5 kHz. For a sim-plified beginning, a very practical analyzercan be built with only one resolution band-width of 300 kHz.

Second Mixer and Second LocalOscillator

Figure 7 shows the second mixer andrelated LO. The heart of this module—andto some extent that of the entire analyzer—is U202, a high-level second mixer. Thismixer is bombarded by large signals thatare as strong or stronger than those at the

front end. Accordingly, the second mixershould have an intercept similar to that ofthe first mixer. This is the usual weak pointin all too many homebrew spectrum analyz-ers—as well as more than a few receivers!The second mixer, U202, uses a +17 dBmlevel Mini-Circuits TUF-1H. This is notthe place for a current-starved telephonecomponent! The second mixer is terminatedin a high-pass/low-pass diplexer followedby an IF amplifier (Q202) biased at 50 mA.This is a critical stage for dynamic range:Don’t replace it with a monolithic substi-tute of reduced gain or intercept.

The second LO begins with a 100 MHz,fifth-overtone crystal oscillator (Q201),followed by a pad and a power amplifier.The oscillator inductor, L201, in Q201’scollector is made of five turns of #22 wirewound on a 6-32 machine screw. (Removethe screw before installing the coil.) Here’san excellent way to align the oscillator:

Temporarily replace the crystal, Y201,with a 51 Ω resistor. Adjust the tuned cir-cuit until oscillation occurs at the desired100 MHz frequency. Then, replace the51 Ω resistor with the 100 MHz crystal; nofurther tuning is required. Measure theoscillator’s output with a power meter be-fore applying it to U202. Adjust the padattenuation (R205, R206, R207) to realizethe specified LO drive level.

After the second LO is operating, attachit to the second mixer and the rest of theanalyzer. With a second mixer input of–35 dBm at 110 MHz, you should obtain areference-level response.

Voltage-Controlled Local Oscillatorand First Mixer

Figure 8 shows the analyzer’s swept LO.The foundation for this module is a Mini-Circuits POS-200 VCO module, U101.Similar VCOs are available from many

August 1998 39

Figure 6—Resolution filters: The upper schematic shows the 300 kHz bandwidth 10 MHz LC filter. If desired, that circuit can berealigned at 10.7 MHz without other design changes. The LC filter is shown as a separate unit connected to the rest of the analyzerwith coaxial cable. However, the filter can be constructed on the board with the crystal filter and relays. Unless otherwise specified,resistors are 1/4 W, 5% tolerance carbon-composition or film units. Equivalent parts can be substituted.

C501, C502, C504, C505, C507, C508,C510, C511, C513, C514, C515—Silvermica or NP0 ceramic, 5%

C503, C506, C509, C512—65 pF plastic

dielectric trim cap (Sprague-GoodmanGYD65000)

K501, K502—SPDT relay; AromatTF2-12V used here (one contact set

unused); values of associated droppingresistors may need adjustment.

L501-L504—17 turns of #22 enameledwire on a T-50-6 toroid (1.15 µH),Q > 250

vendors.10 The VCO output is about+10 dBm, too low a level for the high-levelmixer. A MAV-11 amplifier, U102, pre-ceded by a pad to provide level adjustment,increases the signal level. Confirm the out-put power level before applying it to themixer, U103.

Once the VCO output level is adjustedand confirmed, calibrate its frequencyagainst the VCO control voltage. If a VHFcounter is not available, you can obtain afew points by tuning the VCO to hit localFM broadcast signals of known frequency.Calibrating the VCO is useful if the moduleis used later as a signal source for align-ment of the 110 MHz band-pass filter.

Figure 8 also shows the input mixer,U103, another Mini-Circuits TUF-1H, ter-minated in a 6 dB pad. Although the paddegrades the noise figure, it presents a solidoutput termination for the mixer. Thistermination is reflected, helping to pro-vide a good mixer-input impedance match,important in a measurement instrument.The pad is followed by a Mini-CircuitsMAV-11 IF amplifier (U104) that restoresthe gain lost in the mixer and pad.

The mixer application differs from a

normal diode ring: The RF input is nowattached to the dc-coupled port. This allowsinput frequencies as low as 50 kHz to beconverted to the first IF. The low-frequencyend is limited by mixer LO to RF isolation,which determines the LO energy thatreaches the first IF. The related on-screenresponse is often termed the zero spur, afamiliar “feature” in most RF spectrumanalyzers.

This module (VCO and first mixer) iscontained in a shielded enclosure with co-axial inputs and outputs, including coaxialrouting of the VCO control voltage. Thefront end is susceptible to any VHF andUHF signals reaching it, making shieldingand decoupling especially important.

The 110-MHz IF Band-Pass FilterOne of the more-critical blocks in the

analyzer is the filter that establishes thebandwidth of the VHF IF. The bandwidthmust be at least as wide as the widest10 MHz filter, but must be narrow enoughto reject the 90 MHz second-conversionimages by 80 dB or more. This performanceis only available with a three-pole orhigher-order filter. The best double-tuned

circuits we built (or computer simulated)came close, but just didn’t cut it.

We described double-tuned circuits indetail in a 1991 QST tutorial paper.11 Thosemethods have recently been extended tothree-resonator filters.12 One of the meth-ods presented in the later paper is a sequen-tial approach that begins with a double-tuned circuit (DTC). First, a DTC is builtfor the desired 3 dB bandwidth and has itsperformance confirmed with a widebandsweep (a vital requirement!) Then, a thirdresonator is inserted between the originaltwo. Coupling elements similar to the onethat produced the required DTC bandwidthare repeated in the triple-tuned circuit, butend-section loading is not changed. Thecenter frequency of the three resonators isaligned to complete the filter.

The schematic for the 110 MHz triple-tuned circuit is shown in Figure 9. The in-ductors,100 nH, are made by winding 5turns of #18 wire on the shank of a 1/4-inchdrill bit. These inductors typically have anunloaded Q of just over 200 at 110 MHz.Larger-diameter inductors would have pro-duced higher unloaded Q with the atten-dant lower insertion loss. However, the

40 August 1998

Figure 7Second mixer and second LO.L203 in the output of U202 consists of 17turns of #28 enameled wire on a T-30-6toroid. The actual value is not critical anda molded RF choke can be used in placeof the toroid. Unless otherwise specified,resistors are 1/4 W, 5% tolerance carbon-composition or film units. Equivalent partscan be substituted.C201, C202—470 pF ceramicC203, C206—15 pF NP0 ceramic or silver micaC204, C207—82 pF NP0 ceramic or silver micaC205—65 pF plastic dielectric trim cap (Sprague-Goodman GYD65000)C211—120 pF, silver mica or NP0 ceramicL201—57 nH; 5 turns of #22 wire wound on a 6-32 machine screw. Remove the screw before installing the coil.L202—1 µH molded RFC; any value from 100 nH to 2.7 µH is okayL203—0.3 µH; 9 turns #24 enameled wire on a T-30-6 coreQ201—2N5179Q202—2N5109, 2N3866, 2SC1252, etcU201—Mini-Circuits MAV-11U202—Mini-Circuits TUF-1H mixerT201—10 bifilar turns #28 on FT-37-43 ferrite toroid

stray coupling between coils would haveincreased, which would have necessitatedshields between filter sections. The smaller(10 nH) end-matching inductors are one-inch lengths of #18 wire. The triple-tunedfilter, and its parent DTC, have bandwidthsof 2 to 3 MHz.

The filter alignment and experimenta-tion is usually done with a sensitive powermeter,13 a step attenuator and a signalsource. As mentioned earlier, the VCO canserve the role of signal generator, if one isnot available.

The second-conversion image rejectionis easily measured with a finished analyzer.Apply a 40 MHz signal to the analyzer andadjust it for a reference-level response.Don’t touch the analyzer tuning, but movethe signal generator to 60 MHz. An imagesignal may appear at the same point onscreen as the original 40 MHz signal. Therejection was only 66 dB with a DTC usedfor some experiments. The triple-tuned fil-ter produced 90-dB rejection. The slightextra effort of the triple-tuned circuit iseasily justified. No PC board is availablefor the IF filter.

Input Low-Pass FilterA 70 MHz low-pass filter is shown in

Figure 10. This circuit and a step attenuatorare housed in separate shielded enclosuresin one of our analyzers. In the other, the fil-ter and the attenuator remain outboard ele-ments. Integral components are more con-venient for routine analyzer applications,but incorporation removes them from theequipment pool available for other experi-ments. Also, operation without an outboardlow-pass filter allows the instrument to be

August 1998 41

Figure 8—Front-end for the spectrum analyzer. The LO drive level is set to between +16 and +18 dBm by trimming the padattenuation. The 1 µH inductors used in the MAV-11 outputs can be molded RF chokes or made of 17 turns of #28 enameled wirewound on T-30-6 toroids. Unless otherwise specified, resistors are 1/4 W, 5% tolerance carbon-composition or film units. Equivalentparts can be substituted.L101, L102—1 µH molded RFC; any value between 100 nH to 2.7 µH suitable

U101—Mini-Circuits POS-200 VCOU102, U104—Mini-Circuits MAV-11

U103—Mini-Circuits TUF-1H

Figure 9—VHF band-pass filter used in the 110 MHz IF.

42 August 1998

Figure 10—Input 70 MHz low-pass filter. The filter started as a ninth-orderChebyshev design, but was modified through computer manipulation to use equal-value inductors and standard-value capacitors. The inductors each consist of 8 turnsof #22 wire. The wire is wound on the threads a 1/4-20 bolt as a form; remove the boltafter winding the inductor. See text.

used well into the low UHF area by operat-ing the mixer with VCO harmonics. (Nocircuit boards are available for this filter.)

Construction and Adjustment HintsThe spectrum analyzer can be built using

a number of RF techniques. Our analyzersare collections of small boxes usingcoaxial-cable interconnections. Power sup-plies reach the interior of RF modulesthrough feedthrough capacitors. The onlyopen boards in the system are the time baseand (in one instrument) the log amp, bothconstrained to low frequencies. The110 MHz IF filter and all RF circuit boardsare designed to fit in Hammond 1590Bdiecast aluminum alloy boxes.

The sensitivity of such RF measurementequipment justifies the extensive shielding.Spurious responses are readily seen on thedisplay. While a few can be tolerated, theyseriously detract from a measurementwhen, for example, spurious transmitterproducts are being examined.

Most of the adjustment has already beendiscussed. Indeed, by the time the analyzeris finished, there is little alignment left! Theblock diagram (Figure 1) has typical signallevels shown in parentheses.

The finished analyzer is set up with a–30 dBm signal from a stable-impedancesource. Adjust the IF gain to establish thereference level. If the 10 dB per divisiontracking is not quite correct, change the logamp gain, followed by a readjustment ofthe IF gain until tracking is correct. IF gainis adjusted to the reference level wheneverthe analyzer is used.

Next month, we’ll present methods toextend the analyzer to higher frequencies.We’ll discuss simple test equipment thatcan be used in alignment and present sometypical examples of spectrum-analyzer use.

Notes1Homebrew analyzers described in amateur lit-

erature in recent years have often used TVtuners as the front end. See, for example, AlHelfrick, K2BLA, “An Inexpensive SpectrumAnalyzer for the Radio Amateur,” QST, Nov1985, pp 23-29. A more recent example isFred Brown, W6HPH, “Build a 5- to 850-MHzSpectrum Analyzer,” Communications Quar-terly, Winter, 1997, pp 91-96.

2The usual unit for power measurement with aspectrum analyzer is dBm. This is an abso-lute, impedance invariant unit, signifyingpower compared with 1 mW. Hence, –30 dBmis 30 dB below a milliwatt, or a microwatt. ThedBm unit is especially useful in spectrumanalysis where two levels are often com-pared. The difference between two powers

in dBm is a power ratio in dB. (See JayCraswell, WBØVNE, “Converting BetweendBm, Milliwatts and Watts,” Technical Corre-spondence, QST, Jul 1998, p 70.—Ed.)

3Excellent homebrew VCOs can be builtusing the work of Allan Victor, WA4MGX,“Wideband VCO Design,” Ham Radio Maga-zine, Jul, 1984, pp 49-58. See also the recentwork of Colin Horrabin, G3SBI, as reprintedin “Tech Notes,” Communications Quarterly,Winter, 1996, pp 94-104.

4Many amateurs are uncomfortable with op-amp circuits, a situation easily remedied by astudy of the excellent text by Horowitz andHill, “The Art of Electronics,” 2nd Edition,Cambridge University Press, 1989. See thegolden rules presented on page 177. An op-amp integrator is presented on page 222. Theinformation on sawtooth oscillators beginningon page 288 is also useful.

5PC boards for several of the analyzer circuitsare available from Kanga USA. Details areavailable at the Kanga Web site, http://www.bright.net/~kanga/kanga/ , or fromKanga USA, 3521 Spring Lake Dr, Findlay,OH 45840.

6The best digital audio recording and playbacksystems have a dynamic range around 90 dB.

7An outstanding fundamental treatment of logamps is given in applications data from Ana-log Devices. See, for example, the data sheetfor the AD8307, Analog Devices, Norwood,MA, 1997, http://www.analog.com/prod-uct/Product_Center.html .

8The log/IF board available from Kanga (seeNote 5) is configured for the MC3356. Mostof the 20 pins allotted to the IC are not usedand their locations can be employed to bread-board and retrofit the AD8307 (8 pin DIP) intothe system, if desired.

9If crystal filters are built especially for thisproject, we recommend a peaked responseshape (such as Gaussian-to-6 dB). We haveinvestigated wider bandwidth crystal filters aspresented in Wes Hayward, W7ZOI, “Refine-ments in Crystal Ladder Filter Design,” QEX,Jun, 1995, pp 16-21. See also the carefulwork of Bill Carver, “High Performance Crys-tal Filter Design,” Communications Quarterly,Winter, 1993, pp 11-18, and that of JacobMakhinson, N6NWP, “Designing and Build-ing High Performance Crystal Ladder Filters,”QEX, Jan 1995, pp 3-17.

10See the Mini-Circuits catalogs and applica-tions manuals. A family of viable VCOs arealso available from Synergy Microwave.These units are more expensive, but offerlower phase noise, which would be signifi-cant for more-stringent applications.

11Wes Hayward, W7ZOI, “The Double TunedCircuit: An Experimenter’s Tutorial,” QST,Dec 91, pp 29-34.

12Wes Hayward, W7ZOI, “Extending theDouble-Tuned Circuit to Three Resonators,”QEX, March-April, 1998, pp 41-46.

13Denton Bramwell, K7OWJ, “The Microwatter,”QST, Jun 1997, pp 33-35. Also, see thesidebar in the referent of Note 11.

Wes Hayward, W7ZOI, and Terry White,K7TAU, are not strangers to QST. This is alsonot the first QST project on which they’veworked together. The first, “The Mountain-eer—An Ultraportable CW Station,” describeda QRP transceiver and appeared in the August1972 issue. At the time this spectrum-analyzerproject evolved, the authors worked together inthe Advanced Circuits section of the ReceiverGroup at TriQuint Semiconductor in Hillsboro,Oregon. Over the years, Wes has providedreaders of QST, The Handbook and otherARRL publications with a wealth of projectsand technological know-how. You can contactWes at 7700 SW Danielle Ave, Beaverton, OR97008; e-mail w7zoi@teleport.com. Terrycan be reached at 9480 S Gribble Rd, Canby,OR 97013; e-mail twhite@tqs.com.

New ProductsATV TRANSMITTER FROMPC ELECTRONICS◊ The TX70-10 ATV transmitter from PCelectronics is aimed at operators who usecable-ready TVs (tuned to channels 57-60)to receive amateur television signals in the70-cm ham band. The continuous-duty trans-mitter puts out 10 W AM (color or black-and-white, just like broadcast TV stations)and features a built-in T/R relay, variablepower output, front-panel RCA jacks forvideo and audio (with line and mic inputs), avideo monitor circuit and a PTL jack (push-to-look) in parallel with the unit’s T/Rswitch.

The TX70-10 is housed in a die cast alu-minum box measuring about 7×5×3 inchesand is powered by 13.8 V dc at 3 A. The unitcomes with one user-specified crystal in the440-MHz range. An optional second front-panel-selected crystal is available for $20.A repeater version, the RTX70-10, is alsoavailable.

Price: $439 shipped to the lower 48 states.For more information, contact PC Electron-ics, 2522 Paxson Ln, Arcadia, CA 91007; tel626-447-4565; tomsmb@aol.com.

August 1998 43

C701-C705—NP0 ceramic, 5% tolerance. L701-L704—8 turns #22 bare wire woundin 1/4-20 bolt threads; see text.

September 1998 37

InPart 1,14 we described thedesign and construction of asimple, yet useful spectrumanalyzer. This installment pre-

sents some applications and methods thatextend the underlying concepts.

Amplifier Gain EvaluationOne use for a spectrum analyzer is ampli-

fier evaluation. We can illustrate this with asmall amplifier from the test-equipmentdrawer—an old module that has been pressedinto service for a variety of experiments. Thiscircuit, shown in Figure 11, is used for illus-tration only and is not presented as an opti-mum design. It’s a project that grew fromavailable parts and may be familiar to somereaders. The circuit uses four identical2N5179 amplifier stages. A combination ofemitter degeneration and parallel feedbackprovides the negative feedback needed tostabilize gain and impedance. (Ideally, con-struction and measurement of a single stageshould precede construction of the completeamplifier.)

We began the experiment by setting thesignal generator at a known power level,–20 dBm. (If a good signal generator is un-available, you can easily build a suitablesubstitute.15 We used a surplus HP8654A formost of these experiments.) With the genera-tor and spectrum analyzer connected to eachother through 50-Ω coax cable, we set thegenerator to 14 MHz with 10-dB attenuationahead of the analyzer. The amplifier was notyet connected. We adjusted the analyzer’s IFGAIN control for a response at the top ofthe screen. Using a resolution bandwidth of300 kHz allows for a fast sweep without dis-tortion. In our case, the second harmonic wasonly 26 dB below the peak response. How-ever, when we added a 15 MHz low-passfilter, the second harmonic dropped to–57 dBc.16 The third harmonic is well into the noise. (It is not unusual for a signal gen-erator to be moderately rich in harmonics.)

Next, we inserted the amplifier betweenthe signal generator and the analyzer, keep-ing the low-pass filter in the generator out-put. The on-screen signal went well abovethe top as soon as the amplifier was turnedon. Decreasing the signal generator output to–51 dBm produced the same –20 dBm ana-

By Wes Hayward, W7ZOI, and Terry White, K7TAU

A Spectrum Analyzer forthe Radio Amateur—Part 2

14Notes appear on page 40.

Here aresomeexamples ofproceduresyou can useto becomefamiliar withyour newspectrumanalyzer.

lyzer signal that we saw before the amplifierwas inserted. The gain measured 31 dB.17 In-creasing the analyzer attenuation by 10 dB(for a reference level of –10 dBm) and in-creasing the generator output to –41 dBmproduced the same gain, but a growing har-monic output. The second-harmonic re-sponse was now at –43 dBc and a third har-monic appeared out of the noise at –60 dBc.

We continued the process—moving bothstep attenuators produced an amplifier out-put of 0 dBm with second and third har-monics at –28 and –36 dBc, respectively.The next 10-dB step, however, didn’t workas well, producing gain compression. Witha drive of –21 dBm, the output was only+4 dBm, a gain of only 25 dB instead of thesmall-signal value of 31 dB.

Amplifier Intermodulation DistortionNext, we measured intermodulation dis-

tortion (IMD). The setup for these experi-ments is shown in Figure 12. Two crystal-controlled sources18 at 14.04 and 14.32 MHzare combined in a 6 dB hybrid combiner (re-turn-loss bridge),19 applied to the 15 MHzlow-pass filter and a step attenuator having1 dB steps. This composite signal drove theamplifier, with its output routed to the spec-trum analyzer. For this measurement, wedropped the resolution bandwidth to 30 kHz.

(The video filter was turned on and the sweeprate reduced until the signal amplitude wasstable.) The analyzer’s attenuator, set for areference level of –10 dBm at the top of thescreen, was confirmed with a calibration sig-nal from the signal generator. We adjustedthe IF GAIN to compensate for changes inanalyzer bandwidth and for log-amplifierdrift.

The output of the two-tone generator sys-tem was adjusted to produce a spectrum ana-lyzer response of –10 dBm per tone. TheIMD responses were readily seen, now 47 dBbelow the desired output tones. The outputintercept is given by

IP PIMDR

out out32

= +

where Pout is the output power of each de-

sired tone (–10 dBm) and IMDR is theintermodulation distortion ratio, here 47 dB.The output intercept for this amplifier mea-sured +13.5 dBm. This is well in line withexpectations for such a design.

When performing IMD measurements,it’s a good idea to change the signal levelwhile noting the resultant performance.Dropping the drive power by 2 dB shouldcause a –2 dB response in the desiredtones, accompanied by a 6 dB drop in distor-tion-product tones. The output intercept

should remain unchanged.The IMD in the preceding example was

47 dB below the desired output tones, a valuethat we obtained by simply reading it fromthe face of the ’scope, possible because weuse a log amplifier that has moderate log fi-delity. If the log amplifier was not as accu-rate as it is, we could still get good measure-ments. In this example, you would note thelocation of the distortion products on the dis-play. Then, using the step attenuator, de-crease the desired tones until they are at thenoted level. The result would be –47 dBc forthe distortion level, a measurement that de-pends solely on the accuracy of the attenua-tor. This illustrates the profound utility of agood step attenuator, an instrument that canbe the cornerstone of an excellent basementRF laboratory.

During the third-order output-interceptdetermination just described, we assumedthat the distortion was a characteristic of theamplifier under test. This may not be true. Itis important to determine the IMD character-istics of the spectrum analyzer used for themeasurements before the amplifier measure-ments are fully validated. Specifically, forresults to be valid, the input intercept of theanalyzer should be much greater than theoutput intercept of the amplifier under test.

The spectrum analyzer input intercept iseasily measured with the same equipmentused to evaluate the amplifier. The two tonesare applied to the analyzer input with no at-tenuation present at the analyzer front end.Then, the input tones are adjusted for a full-screen response. In this condition, thereshould be no trace of distortion. Althoughthis is generally an adequate test, it does notestablish a value for the input intercept. Todo that, we must overdrive the analyzer, us-ing signals that exceed the top of the screen.

The following steps were used to measurethe analyzer input intercept:

• We calibrated the analyzer for a referencelevel of –30 dBm with a 30-kHz resolution.

Figure 11—Sample wideband amplifier used to illustrate amplifier measurements.

Figure 12—Equipment setup for evaluation of amplifier intermodulation distortion.

Figure 13—Test setup for impedance-match measurements with return-lossbridge.

• Confirmed the lack of on-screen distor-tion with two tones at the reference level.

• Increased the drive of each tone by10 dB to provide a pair of –20 dBm tones tothe analyzer. This higher-than-reference-level input produced distortion products66 dB below the reference level, or –96 dBm.The input signals producing this were each–20 dBm, so the IMD ratio is (–20) – (–96) =76 dB. Following the earlier equation, theinput intercept was +18 dBm.

• A 2-dB drive increase produced the ex-pected 6-dB distortion increase. If this hadnot occurred, distortion measurements underoverdrive would be suspect. The +18-dBmvalue seems to be a good number. Thisanalyzer generally seems happy with signals20 dB above the top of the screen, but notmuch more.

The intercept for the analyzer with attenu-ation in place is the measured value with nopad plus the attenuation. Hence, with 20 dBof attenuation, the input intercept will be +38 dBm, and so forth.

Return-Loss MeasurementsThe next amplifier characteristic that we

measured was the input impedance match, orreturn loss, performed with the setup shownin Figure 13. With the signal generator set fora 14-MHz output of about –30 dBm, we set

the analyzer for full-scale response with theLOAD port of the return-loss bridge open cir-cuited. Placing a 50-Ω termination momen-tarily on the LOAD port, produced a 38-dBsignal drop. This is a measure of the bridgedirectivity. A 38-dB directivity is more thanadequate for casual measurements.

Then, we removed the termination from

38 September 1998

we want to experiment on the VHF and UHFbands, but we need to examine higher-orderharmonics of HF gear. One method we canuse with a regular receiver is a converter,usually crystal controlled. The same can bedone with a spectrum analyzer, althoughcrystal control is not needed. We can build asimple block converter, consisting of noth-ing more than a 100 to 200 MHz VCO (justlike that used in the analyzer) and a diodering mixer. A Mini-Circuits POS-200 VCOwith a 3 dB pad will directly drive a Mini-Circuits SBL-1 or TUF-1 mixer to produce ablock converter with a nominal loss of 10 dB.(One of the spectrum analyzer front-endboards could be used, with slight modifica-tion, for the block converter.)

This block converter allows analysis ofmuch of the VHF spectrum. With the con-verter VCO set at 100 MHz, frequencies from100 to 170 MHz are easily studied. The 70 to100-MHz image is also available—it can alsolead to confusion, as a few minutes with asignal generator will demonstrate. With theconverter VCO up at 200 MHz, the 200 to270 MHz spectrum is also available. Clearly,there is nothing special about the particularVCO used in the converter. All that is requiredto convert other portions of the low UHFspectrum to the analyzer range is a differentVCO, and perhaps, a higher-frequency mixer.We will soon build similar block convertersto allow analysis of the 432-MHz area. Oneof the popular little UHF frequency countersis quite useful with these converters. Theblock converters should be well shieldedand decoupled from the power supply.

Even without the analyzer, the block con-verter is a useful tool. For example, it can beused with a 10 MHz LC band-pass filter andan amplifier to directly drive a 50-Ω-termi-nated oscilloscope. This can serve as a sensi-tive detector for the alignment of a 110-MHzfilter.

As useful as the block downconverter is, ithas image-response problems that can greatlyconfuse the results. The preferred way to getto the higher frequencies is with a spectrumanalyzer—just like the one we have de-scribed—but having higher-frequency oscil-lators and front-end filters. If you’re careful,you can build helical resonator filters for the500-MHz region, or higher, with sufficientlynarrow selectivity to allow a second conver-sion to 10 MHz. A more practical route uses athird IF. Such a triple-conversion analyzer isshown in Figure 17, where a 1 to 1.8 GHzVCO moves signals in the 0 to 800-MHzspectrum to a 1 GHz IF. This signal is stilleasily amplified with available monolithicamplifiers. Building a 1 GHz filter should notbe too difficult, for a narrow bandwidth is notrequired. A bandwidth of 20 or 30 MHz witha three-resonator filter would be adequate.The resulting signal is then heterodyned to aVHF IF (such as 110 MHz) where the remain-ing circuitry is now familiar.

Clearly, there are several ways to attackthe project. Recent technology offers somehelp in the way of interesting band-pass filterstructures, as well as high-performance,low-noise VCOs. Time spent with some cata-

Figure 14—Output of a typical QRP trans-ceiver kit. The 1-W plus output at 7 MHz isthe dominant signal; all others arespurious outputs more than 40 dB down,currently meeting FCC specifications.

Figure 15—Output of a simple homemadeQRP rig. The desired signal is the largepip at the center of the trace. Twomeasurable spurious responses exist, oneto the left and one to the right of the mainsignal. The response to the left is 3.5 MHzfeedthrough from the VFO at –64 dBc; theresponse to the right of the desired signalis a second harmonic at –44 dBc.

level low enough that the amplifier remainslinear. In this case, we saw no difference inthe results with drive that was 10 dB higher.

When performing return-loss measure-ments—and indeed, most spectrum-analyzermeasurements—it is wise to place at least10 dB of attenuation ahead of the spectrum-analyzer mixer. When this attenuation isswitched in, the reference level changes fromthe top of the screen to a point down screen abit. Return loss is then measured as a decibeldifference with regard to the new reference.

Antenna MeasurementsIt is interesting to look at some other

impedance values while the return-lossbridge is attached to the signal generator andspectrum analyzer. The obvious choice is thestation antenna system, especially if it isconnected through a Transmatch. Playingwith the tuning will readily demonstrate thatthe return-loss bridge and sensitive detec-tion system will allow adjustments to accu-racy unheard of with traditional diode detec-tor systems. Although such tuning accuracyis not needed in a normal antenna installa-tion, it is interesting to see what can be mea-sured when the need does arise.

Transmitter EvaluationAnother obvious application for a spec-

trum analyzer is in transmitter evaluation.Figure 14 shows the output of a typical QRPtransceiver kit. The 1-W plus output at 7 MHzis the dominant signal, with all others beingtransmitter spurious outputs. All spurs aremore than 40 dB down, which meets currentFCC specifications. On the other hand, signi-ficantly better performance is easily obtained,especially if the builder has the facilities tomeasure them. Figure 15 is a photograph ofa simpler QRP rig with two measurable spu-rious responses. One is the 3.5-MHz feed-through from the VFO at –64 dBc; the secondis a harmonic at –44 dBc.

The output available from a typical QRPrig (and certainly higher power rigs) isenough to damage the spectrum-analyzerinput circuits. Attenuators that we generallybuild are capable of handling 0.5 to 1 Winput without damage, while commercialattenuators are rated at from 0.5 to 2 W in-put. The mixers used in this analyzer can bedamaged with as little as 50 to 100 mW sig-nals. Two methods can be employed to viewthe output of a high-power transmitter with-out causing damage to the spectrum ana-lyzer. In one, the transmitter output is runthrough a directional coupler with weak cou-pling to the sampling port—perhaps –20 to–30 dB. The majority of the output is dissi-pated in a dummy load. The second methoduses a fixed, high-power attenuator. Figure16 shows an attenuator that will handleabout 20 W while providing 20-dB attenua-tion. The design is not symmetrical.

Spectrum Analysis at HigherFrequencies

Although the 70-MHz spectrum analyzeris extremely useful, we constantly wish thatit covered higher frequencies. Not only do

the bridge and placed it on the amplifier out-put. A short length of cable connected thebridge load port to the amplifier input withpower applied to the amplifier. The resultwas a response 20 dB below the top of thedisplay; 20 dB is the return loss for the am-plifier input, an excellent match for a gen-eral-purpose amplifier.20 The match im-proved slightly when the output load wasremoved, an unusual situation.

The next variation measured the outputmatch for the amplifier. The load was trans-ferred to the amplifier input and the cable fromthe bridge LOAD port was moved to the ampli-fier output, producing a reading 15 dB belowthe screen top. This match was virtually un-changed when the input termination was re-moved. The weak dependence of both ampli-fier return losses is the result of a four-stagedesign. A single-stage feedback amplifier willhave port impedances that depend strongly onthe termination at the opposite port.

The –30 dBm drive from the generatorprovides an available power of –36 dBm fromthe bridge LOAD port. This is low enoughthat the amplifier is not over-driven. Thematch measurements should be done at a

September 1998 39

Figure 16—A 20-W, 20-dB attenuator for transmitter measurements.

logs and a Web browser could be very pro-ductive in this regard. Just as important is theexperience that the lower-frequency analyzerwill provide. Not only is it the tool (perhapssupplemented with block converters) that isneeded to build a higher-frequency spectrumanalyzer, but it is the vehicle to provide theconfidence needed to tackle such a chore.

SummaryThis analyzer has been a useful tool for

over 10 years now. It was a wonderful ex-perience and fun to build HF and VHF CWand SSB gear with the “right” test equip-ment available. We have built special, nar-row-tuning-range analyzers to examinetransmitter sideband suppression and distor-tion. The equipment uses the same conceptspresented.

There are many ways that this instrumentcan grow. One builder has already bread-boarded a tracking generator. (Let’s see thatin QST!—Ed.) Many builders will want tointerface the analyzer with a computer in-stead of an oscilloscope. A recent QST net-work analyzer paper suggests circuits thatmay provide such a solution.21

A recent QST summary of WRC9722 out-lines new specifications regarding spuriousemissions from amateur transmitters. Gener-

ally, the casual specifications that we haveenjoyed for many years are being replaced bynew ones that are more stringent—and morerealistic in safeguarding the spectral envi-ronment and reflecting the sound designs thatwe all strive to achieve. Equipment such asthe spectrum analyzer described here canprovide the basic tool needed to meet thisnew challenge.

AcknowledgmentsMany experimenters had a hand in this

project and we owe them our gratitude. JeffDamm, WA7MLH, and Kurt Knoblock,WK7Q, built versions of the analyzer andhave garnered several years of use with them.Their experiences have been of great value inour efforts. Barrie Gilbert of Analog DevicesNorthwest Labs suggested the AD8307 logamplifier and provided samples and earlydata needed for evaluation measurements.Many of our colleagues within the WirelessCommunications Division at TriQuint Semi-conductor have helped us with filter mea-surements: Thanks go to George Steen and toDon Knotts, W7HJS. Finally, special thanksgo to our colleague in the receiver group atTriQuint, Rick Campbell, KK7B, who pro-vided numerous enlightening discussionsand suggestions regarding the preparation of

the paper and the role of measurements inamateur experiments.

Notes14Wes Hayward, W7ZOI, and Terry White,

K7TAU, “A Spectrum Analyzer for the RadioAmateur,” Part 1, QST, Aug 1998, pp 35-43.

15The wide-range oscillator presented in Fig 68,Chapter 7 of Solid-State Design for the RadioAmateur (Newington: ARRL, 1997) is still in-tact and still often used.

16The term dBc refers to dB attenuation withrespect to a specific carrier.

17Formally, this is the transducer gain, or 50-Ωinsertion gain. There are many different pa-rameters that are called “gain.”

18The signal sources used are updated ver-sions of the circuits shown in Fig 66, p 168, inthe Note 15 referent.

19See page 154, Solid-State Design for theRadio Amateur.

20A 20-dB return loss corresponds to a voltagereflection coefficient of 0.1, or an SWR of1.222. See Wes Hayward, W7ZOI, Introduc-tion to RF Design (Newington: ARRL, 1994),p 120.

21See, for example, Steven Hageman, “BuildYour Own Network Analyzer,” QST, Jan1998, pp 39-45; Part 2, Feb 1998, pp 35-39.

22Larry Price, W4RA, and Paul Rinaldo, W4RI,“WRC97, An Amateur Radio Perspective,”QST, Feb 1998, pp 31-34. See also RickCampbell, KK7B, “Unwanted EmissionsComments,” Technical Correspondence,QST, Jun 1998, pp 61-62.

An inside view of a prototype VHF band-pass filter. See Figure 7.

Figure 17—Blockdiagram for a version ofthis analyzer that willfunction into the VHF andlow-UHF regions.

40 September 1998

74 November 1999

W1AWSchedule

Feedback

Yaesu FT-90R Dual-Band FM MobileTransceiver (see: “Product Review,” Sep-tember 1999 QST). Minimum Bid: $275.

Yaesu FT-100 MF/HF/VHF/UHF All-Mode Transceiver with XF-117C 500-HzCW filter, S/N 9D021081 (see: “ProductReview,” June 1999 QST). Minimum Bid:$995.

Yaesu FT-100 MF/HF/VHF/UHF All-Mode Transceiver with XF-117C 500-HzCW filter, S/N 9D021145 (see: “ProductReview,” June 1999 QST). Minimum Bid:$995.

Yaesu VX-5R Tri-Band 6m/2m/70cm FMHand-Held Transceiver (see: “Product Re-view,” May 1999 QST). Minimum Bid:$155.

Sealed bids must be submitted by mailand must be postmarked on or before De-cember 1, 1999. Bids postmarked after theclosing date will not be considered. Bidswill be opened seven days after the closingpostmark date. In the case of equal high bids,the high bid bearing the earliest postmarkwill be declared the successful bidder.

In your bid, clearly identify the item youare bidding on, using the manufacturer’sname and model number, or other identifi-cation number, if specified. Each item re-quires a separate bid and envelope. Ship-ping charges will be paid by ARRL. Pleaseinclude a daytime telephone number. The

◊ Please refer to Francesco Morgantini,IK3OIL, “A PIC 16F84-Based CW De-coder,” QST, Aug 1999, pp 37-40. An up-dated program and related files

successful bidder will be advised by tele-phone or by mail. Once notified, confirma-tion from the successful bidder of intent topurchase the item must be made within twoweeks. No response within this period willbe interpreted as an indication of the win-ning bidder’s refusal to complete the trans-action. The next highest bidder will thenhave the option of purchasing the item. Noother notifications will be made, and noinformation will be given to anyone otherthan successful bidders regarding finalprice or identity of the successful bidder. Ifyou include a self-addressed, stamped post-card with your bid and you are not the highbidder on that item, we will return the post-card to you when the unit has been shippedto the successful bidder.

Please send bids to Bob Boucher, Prod-uct Review Bids, ARRL, 225 Main St,Newington, CT 06111-1494.

(IK3OIL2.ZIP) has been posted at theARRL’s download site. Read the includedREADME.TXT file for an explanation ofthe contents and use of the .ASM and .HEXfiles that apply to various LCDs, includingthe Optrex 16117A. Contact FAR Circuitsfor PC board or IC replacements (18N640Field Ct, Dundee, IL 60118-9269; tel 847-836-9148 (voice and fax).

◊ In “Using the MFJ-259 SWR Analyzer toFind a Short Circuit in Coaxial Cable,”Technical Correspondence, QST, Oct 1999,p 66, the unit of measurement for the ShortLocation in Equation 1 should be in feet,not inches.

◊ Sharp-eyed bit masters detected some “par-ity errors” in “Test Your Knowledge” in theOctober 1999 QST, page 52. The final threerows of the truth table in answer 3 shouldshow Amp Key as logic zero. The correctanswer to question 5 is that PTT B should beset to logic zero, lest the amp be keyed when-ever the antennas are ready. The answer forquestion 7 is actually one-half of 33.3 MHzor 16.67 MHz. Thanks to K5RA, N1XS, andWB6IKJ and others.

◊ Please refer to Wes Hayward, W7ZOI, andTerry White, K7TAU, “A Spectrum Ana-lyzer for the Radio Amateur—Part 2, QST,Sep 1998, p 40, Fig 16. The six 2-W inputresistors for the 20-dB pad should be 620 Ωunits, not 820 Ω as originally specified.—Wes Hayward, W7ZOI (tnx EA2SN)

W1AW’s schedule is at the same local time throughout the year. The scheduleaccording to your local time will change if your local time does not have sea-sonal adjustments that are made at the same time as North American timechanges between standard time and daylight time. From the first Sunday inApril to the last Sunday in October, UTC = Eastern Time + 4 hours. For the restof the year, UTC = Eastern Time + 5 hours.

Morse code transmissions:Frequencies are 1.818, 3.5815, 7.0475, 14.0475, 18.0975, 21.0675, 28.0675and 147.555 MHz.Slow Code = practice sent at 5, 71/2, 10, 13 and 15 wpm.Fast Code = practice sent at 35, 30, 25, 20, 15, 13 and 10 wpm.Code practice text is from the pages of QST. The source is given at the beginning

W1AW SCHEDULEPacific Mtn Cent East Mon Tue Wed Thu Fri

6 AM 7 AM 8 AM 9 AM Fast Slow Fast SlowCode Code Code Code

7 AM- 8 AM- 9 AM- 10 AM-1 PM 2 PM 3 PM 4 PM

1 PM 2 PM 3 PM 4 PM Fast Slow Fast Slow FastCode Code Code Code Code

2 PM 3 PM 4 PM 5 PM Code Bulletin

3 PM 4 PM 5 PM 6 PM Teleprinter Bulletin

4 PM 5 PM 6 PM 7 PM Slow Fast Slow Fast SlowCode Code Code Code Code

5 PM 6 PM 7 PM 8 PM Code Bulletin

6 PM 7 PM 8 PM 9 PM Teleprinter Bulletin

645 PM 745 PM 845 PM 945 PM Voice Bulletin

7 PM 8 PM 9 PM 10 PM Fast Slow Fast Slow FastCode Code Code Code Code

8 PM 9 PM 10 PM 11 PM Code Bulletin

Visiting Operator Time

of each practice session and alternate speeds within each session. For example,“Text is from July 1992 QST, pages 9 and 81,” indicates that the plain text is fromthe article on page 9 and mixed number/letter groups are from page 81.Code bulletins are sent at 18 wpm.W1AW qualifying runs are sent on the same frequencies as the Morse codetransmissions. West Coast qualifying runs are transmitted on approximately3.590 MHz by W6OWP, with K6YR as an alternate. At the beginning of eachcode practice session, the schedule for the next qualifying run is presented.Underline one minute of the highest speed you copied, certify that your copywas made without aid, and send it to ARRL for grading. Please include yourname, call sign (if any) and complete mailing address. Send a 9×12-inch SASEfor a certificate, or a business-size SASE for an endorsement.

Teleprinter transmissions:Frequencies are 3.625, 7.095, 14.095, 18.1025, 21.095, 28.095 and147.555 MHz.Bulletins are sent at 45.45-baud Baudot and 100-baud AMTOR, FEC Mode B.110-baud ASCII will be sent only as time allows.On Tuesdays and Fridays at 6:30 PM Eastern Time, Keplerian elements formany amateur satellites are sent on the regular teleprinter frequencies.

Voice transmissions:Frequencies are 1.855, 3.99, 7.29, 14.29, 18.16, 21.39, 28.59 and147.555 MHz.

Miscellanea:On Fridays, UTC, a DX bulletin replaces the regular bulletins.W1AW is open to visitors from 10 AM until noon and from 1 PM until 3:45 PMon Monday through Friday. FCC licensed amateurs may operate the stationduring that time. Be sure to bring your current FCC amateur license or aphotocopy.In a communication emergency, monitor W1AW for special bulletins as fol-lows: voice on the hour, teleprinter at 15 minutes past the hour, and CW on thehalf hour.Headquarters and W1AW are closed on New Year’s Day, President’s Day,Good Friday, Memorial Day, Independence Day, Labor Day, Thanksgivingand the following Friday, and Christmas Day.

Visiting Operator Time(12 PM - 1 PM closed for lunch)