A Portable Low Frequency Antenna Analyzer

For those of us who like to play with antennas on the ham bands, one of the handiest tools to have around the shack is an "antenna analyzer". These gadgets combine a signal generator and standing-wave ratio (SWR) sensor in a small battery-powered unit.

Dave, WB6VKH has come up with a way to extend the coverage of one of the most popular antenna analyzers, the MFJ-259, all the way down to 100 kHz, and an article describing his modification is available here. (Use your browser's Back button to return to this article.) A text version with separate PCX images is also available in PKZip format (82 KB).

I tried to do something similar with my Autek RF-1 but didn't have too much success, so I put it aside and decided to build an analyzer specifically for LowFER use. I wanted an analyzer that would operate in the 100 to 300 kHz range and would measure antenna system resonance, system losses and antenna capacitance, as well as the inductance and Q of loading coils and receiving loops. Unlike hams, LowFERs are not usually concerned about standing-wave ratio (SWR), and I didn't incorporate that measurement into the basic design. However, an optional SWR bridge circuit is included in this article for those who wish to add it.

The analyzer can be built into a 6.25 by 3.75 by 2 inch box and will operate from a single 9- volt battery. Figure 1 is a suggested front panel layout for the project. The main tuning dial has three concentric scales which you will have to calibrate. Only the outer (frequency) scale requires any test equipment during calibration, and if necessary you can get by with an AM radio, using techniques described later in this article. The middle and inner dials are for reading inductance or capacitance. Once the frequency dial is calibrated, you can use a calculator or the lookup table given in this article to determine where to put the markings on the L and C scales.

A full-size photo (44 kb) of the front panel is available for reference, and a panel template is also available (18 kb).

Circuit Description -- This analyzer uses a simple approach in which an RF detector measures the output of a signal generator with a source impedance of approximately 100 ohms. The "unknown" is connected across the output of the generator, loading its output, and the meter in the RF detector circuit is calibrated to show the unknown impedance. When connected to a LowFER antenna that is series-tuned with a loading coil, the generator frequency is varied to obtain the minimum voltage reading. This tells you at what frequency the antenna system is resonant, and the meter reading gives the RF resistance at the resonant frequency. To measure inductance, we simply put a known capacitor in series and tune for resonance, and to measure capacitance we do the same thing with a known series inductor.

By the way, for those who already have a modified MFJ-259, the same technique can be used with it to measure inductance and capacitance. In order to get reasonably accurate readings of the series resistance of high-Q circuits, the signal generator needs to have low harmonic content. The series impedance of the circuit at resonance is very low at the fundamental frequency, but not at the harmonics, and this will cause an error in the reading. Another way around the harmonic problem is to use a tuned detector such as a receiver, but the goal was to put everything into one self-contained box and keep the circuit as simple as possible. The circuit in Figure 2 uses a technique that is common to many laboratory function generators. U1A is an integrator, and U2 is a threshold comparator with hysteresis. By feeding the output of the threshold detector back to the integrator input, we get an oscillator that produces a triangle-wave output at pin 1 of U1A and a square-wave output at pin 7 of U2. Potentiometer R5 (with series resistor R6) and capacitor C6 are the primary frequency-determining elements. Trimpot R6 sets the maximum frequency to which the analyzer will tune, when potentiometer R5 is at its minimum resistance setting.

* Figure 2. LF Analyzer Circuit:

The triangle-wave output from U1A is fed through resistor R3 to a diode network which produces soft clipping of the waveform and transforms it into a fairly good approximation of a sine wave. This circuit has an inherently constant output as the frequency is varied. Resistor R9 and capacitor C7 provide a small amount of filtering to get rid of some of the remaining kinks in the waveform. Transistor Q1 is an emitter follower that delivers approximately 1.5 volts peak-to- peak into resistor R10, which is in series with the output terminals.

Rather than using a simple diode rectifier for the detector, this analyzer circuit has a single- balanced mixer detector. The idea is to preserve linearity at very low RF voltages so that we can measure the low series RF resistances we'd like to see in our antenna systems. This "synchronous" type of detector also reduces problems with RF pickup from nearby broadcast stations, which can affect readings on many of the commercial antenna analyzers that cover the HF range. The DC output of the detector is only about 200 millivolts with no load on the output terminals. U1B serves as a meter amplifier which will bring this level up to several volts when the meter gain adjust potentiometer R14 is set for maximum gain.

S2 is a 12-position switch that connects the generator output to the "antenna" terminal either directly or through a known series inductor, capacitor or resistor. There are five resistors (R18 through R22) which are used for calibration. Only one resistor is actually needed to set the meter circuit gain during routine operation, but there were four unused positions on S2 and I figured it was a good place to store the precision resistors used for calibration. The color codes on those darned things are so hard to read that once they go into the parts box, they are essentially lost to the world anyway.

Components -- Although the circuit may work with devices other than the ones specified for U1A and U2, it's best not to deviate from the parts list unless you feel like doing some redesigning on your own. The only other critical components are the reference capacitors and inductors C11-C13 and L1-L3. They should have 5 per cent or better tolerance, and low losses. Diodes D5 and D6 should be matched. You can do this by buying a package of ten 1N34As from Radio Shack, then using an ohmmeter or a digital multimeter with a diode test function to select two diodes that have nearly the same forward voltage drop. A standard 5-volt regulator like the 7805 or 78L05 can be used for U3. See the Power Supply discussion for more comments on the voltage regulator. Potentiometer R5 should have a linear taper rather than an audio taper. Otherwise the low-frequency end of the dial calibration will be more expanded and the high-frequency end will be more compressed than it already is. The potentiometer specified in the parts list has a tolerance of 20 per cent and an unspecified linearity. There are other parts in the catalogs with 10 per cent tolerance, but they cost 5 to 10 times as much, and are still not precise enough so you could use the same dial calibration that I did.

If you don't feel the need to set the low and high-frequency limits of the tuning range to specific values like 100 and 300 kHz, you can use a fixed resistor (nominally 3.9K) in place of trimpot R6, and a fixed capacitor in place of the parallel combination of C6 and C6a. The nominal capacitance of C6 and C6a in parallel should be about 300 pF if the maximum resistance of R5 is exactly 10K. A 270 pF capacitor for C6 and an 8-50 pF trimmer for C6a will be about right. However, if the potentiometer value is off by 20 per cent, C6 may need to be larger or smaller in order to set the low end of the tuning range to exactly 100 kHz. Resistor R15 should be changed to 4.7K if the Mouser 500 uA meter (shown as an alternate on the parts list) is used instead of the 200 uA meter from Dan's Small Parts. This resistor sets the meter sensitivity to approximately 3 volts full scale.

Construction -- The prototype circuit was built on a Radio Shack No. 276-150A "General Purpose IC PC Board" which was mounted on standoffs behind the frequency adjust pot. The battery is mounted to the left of the PC board as viewed from the front of the analyzer. It takes a little effort to get everything to fit on the board, and there is room in the recommended Radio Shack 270- 627 enclosure for a larger PC board if you want to spread things out a little. However, even though this is a fairly low frequency circuit it's a good idea to keep all leads as short as possible. Figure 3 shows the component placement I used. It's difficult to see the parts, especially the resistors that are mounted vertically on the board, but you can get a general idea of where the parts are located.

* Figure 3. Component Locations

Click for large version. (41 kb)

I ran a tinned bus wire from capacitor C10 on the PC board to the "DIRECT" terminal of S2, and used this as the common lead for all the components going to S2. It's easier than mounting all those parts on the PC board, and allows room to keep inductors L1 through L3 fanned out and spaced away from each other. These inductors should also be spaced at least 1/2 inch away from any metal surfaces like the front panel of the enclosure. Because they act like little loop antennas, they can interact with each other, and their inductance and Q can be affected by nearby metal surfaces. When connecting potentiometers R5 and R14, make sure that R5 is wired so that it has maximum resistance (the lowest frequency setting) when the knob is turned fully counter-clockwise, and that R14 has maximum resistance (maximum meter amplifier gain) with the knob fully clockwise. To make a large pointer knob for the frequency and L-C scales, I cut a 3 inch diameter circle from heavy clear plastic material of the type used for product packaging, and fastened it to the bottom of a standard 1-1/4 inch diameter knob with double sticky tape.

The front panel layout and other figures for the project are available in the file library section of the Longwave Web page. There is also a copy of the "artwork" for the front panel layout, which you can scale on your own computer and print out for use as a hole-drilling template. For those without downloading capability, a large SASE to the author will get you a 1:1 scale printout of the front panel. To dress up the project, you can attach the computer-generated layout to the front of the enclosure with contact cement or double sticky tape, and then cover it with adhesive-backed clear laminating film (available at office supply stores) after the dial is calibrated. A "nibbling tool" (Radio Shack # 64-823) is handy for making rectangular holes like the one needed for mounting the meter. There's another option for the meter -- I sometimes install a meter jack in the front panel and mount a little edge-reading meter to the outside of the case with sticky- backed Velcro. This lets me use the same meter, which is usually the most expensive component, for several different projects. If a permanently-mounted meter of the type recommended in the parts list is used, it should be left uninstalled until after the checkout and calibration is completed, so that you can remove the front of the meter to put in your own calibrated scale.

Checkout and calibration -- Pin 1 of U1 is the triangle-wave test point, pin 7 of U2 is a square- wave output with an amplitude of 5 volts p-p, and a 1.5 volt p-p sine-wave output is available at the emitter of Q1. A triangle wave amplitude of approximately 3 volts p-p is required for proper operation of the sine-wave shaping circuit. In the unlikely event (as the airlines say) that a waveform adjustment is necessary, resistors R4 and/or R7 can be varied to set the amplitude, while resistors R1 and R2 affect the waveform symmetry. All of these resistors will also influence the frequency, so if you make any changes to these components it should be done prior to the frequency calibration.

Use a frequency counter or receiver to calibrate the outer (frequency) dial as accurately as possible. The first step, if you have included a trimpot for R6 and trimmer capacitor C6a in the circuit, is to set the high and low frequency limits. First turn the tuning control R5 fully clockwise (to the highest-frequency setting) and adjust R6 for a frequency of 300 kHz. Then turn R5 fully counter-clockwise and adjust C6a for 100 kHz. As mentioned earlier, it may be necessary to change the value of C6 to get to exactly 100 kHz. However, there's nothing sacred about either 100 or 300 kHz, and you can set the tuning limits to any convenient frequencies (or accept what you get with fixed components). The low- and high-limit adjustments are not independent. It may take a couple of iterations to get the high and low frequency settings just where you want them.

If you don't have a frequency counter or a receiver that covers the 100 to 300 kHz range it's possible to calibrate the analyzer by listening for its harmonics on an AM broadcast receiver. You'll need to know what frequency the radio is actually tuned to, and which harmonic you're hearing. An AM radio with digital tuning and a BFO will let you zero beat the harmonics exactly on frequency. It's also possible to beat the harmonics against incoming broadcast stations if you know their frequencies. Sorting out the harmonics may take a little detective work. For example, with the radio tuned to 600 kHz, it will pick up harmonics of 300, 200, 150, 120, 100 and 85.7 kHz. All of these frequencies may fall within the tuning range of the analyzer with the component values given in Figure 2 (especially since the frequency adjust potentiometer R5 has a resistance tolerance of +/- 20 per cent). By connecting a short clip lead to the square-wave output (pin 7 of U2) and keeping the radio a couple of feet from the analyzer, you should find that the odd harmonics are much stronger than the even harmonics. And by checking at more than one frequency, you can sort out harmonics by elimination. Let's say you have the AM radio tuned to 600 kHz and think you're listening to the 5th harmonic of 120 kHz. If you're right, you should find the 7th harmonic at exactly 840 kHz, and none of the other frequencies with harmonics on 600 kHz will also have a harmonic on 840. Once you have definitely established one frequency, it's fairly easy to finish the frequency dial calibration.

After the frequency dial is calibrated, use Table 1 to determine the calibration points for the middle and inner scales, which are used to read capacitance and inductance. To obtain calibration points not given in the table, use the formula for tuned circuits: F = 0.159/SQRT(LC), with L = 6800 uH for the inner scale and L = 2200 uH for the middle scale. The frequency adjust knob and pointer can be removed while you're marking off the values corresponding to the frequencies in Table 1. Before removing the knob, however, mark the pointer position at one or both limits of the knob travel. This makes it a lot easier to get the knob back in the right orientation. I suggest using a colored pencil or pen to make the markings for the middle L-C scale (the scale corresponding to the lower half of Table 1), and the S2 positions that are enclosed in boxes. This will help reduce confusion about which scale you're supposed to read when measuring inductance or capacitance.

* Table 1. L-C Dial Calibration

Both of the meters shown on the parts list use the same type of construction, in which the case is held together by a piece of clear adhesive tape. This makes it easy to take the meter apart and insert a scale with your own calibration for the resistance readings. An alternate approach is to tape a strip of white paper to the outside front of the meter case, using a strip that covers only the top half of the face so the pointer is still visible. Then you can make the markings on the outside without dissecting the meter. To calibrate the meter, connect a jumper between the analyzer's "antenna" and "ground" terminals, and place S2 in the 150-ohm calibrate position. Adjust the meter gain control R14 for a reading of approximately 95 per cent of full scale and make a 150-ohm calibration mark on the meter face. An alternative way to start the calibration procedure is to set the meter gain for a full-scale reading with nothing connected to the "antenna" terminal. This results in a little less resolution for low-resistance measurements, but will keep the meter from "pinning" with high-impedance loads. After you make the initial setting, mark the position of the meter gain control knob. Without changing the gain setting, turn S2 to the other resistance positions and make the corresponding markings on the meter face.

Power Supply -- The analyzer circuit will operate from supply voltages of approximately 6.5 volts to 15 volts. If something other than the low-dropout 2930L05 regulator called for in the circuit diagram is used, the lower limit on the supply voltage may be a little above 7 volts. Current drain is about 30 mA (somewhat less if you use the Hartley oscillator option described below). This is about the same as the current drain of the Autek RF-1, and the unit will work fine on a 9-volt alkaline battery. For extended operation it can be run from a small AC-operated supply. Most plug-in transformer supplies (also known as "wall warts") can supply enough current for this instrument -- as long as the supply has a filtered DC output rather than raw AC. One characteristic of the signal generator circuit shown in Figure 2 is that it won't work at all unless the supply voltage is at least 6.5 volts. If the analyzer seems to be completely dead, check the battery voltage as a first step in trouble-shooting.

Using the Analyzer -- For antenna measurements, we'll first assume that the antenna already has a loading coil that tunes it on or near the LowFER band. To measure the resonant frequency and loss resistance, turn switch S2 to the "DIRECT" position and connect the antenna and ground to the analyzer terminals. Set the meter gain to the position marked during the initial calibration process, and rotate the frequency dial carefully until you get the deepest null in the meter reading. With luck, this will occur at your intended transmitting frequency. The meter reading at the null shows the total loss resistance in your antenna system. This includes loading coil losses, ground losses, dielectric losses in nearby trees, etc. Note: If your loading coil is part of the final amplifier circuit, disconnect the final and connect the analyzer between ground and the "low" side of the coil (the side that has a bypass capacitor to ground). Be sure to disconnect the bypass capacitor also. The reason for this procedure is to turn the antenna system into a simple series L- C circuit. Making a measurement across a set of taps or a coupling link on the loading coil will not give a loss reading that is easy to interpret, because it has been modified by transformer action. Many experimenters use a variable capacitor between the top of the loading coil and ground to fine tune the antenna. This should also be removed to get an accurate measurement of system losses. If it's necessary to tune very far above 200 kHz to find the resonant point without the tuning capacitor, it means that the loading coil isn't big enough, and you might get a substantial improvement in radiated power by using the proper inductor.

To determine the capacitance of an antenna that doesn't have a loading coil, set S2 to one of the C X 1 positions. A 15-meter high antenna consisting only of a piece of # 14 wire has a capacitance of about 95 pF, which can be read on the inner dial scale with S2 in the unboxed C X 1 position. Near the other extreme, a 25-foot high vertical with twelve 25 foot by 1-inch radials and a # 8 skirt wire should have a capacitance of about 570 pF, which would be read on the middle dial scale with S2 in the boxed C X 1 position. Tune the analyzer for a null and read the capacitance value from the appropriate scale. With no loading coil in the antenna circuit, the antenna capacitance is nearly independent of frequency (as long as the antenna length is not an appreciable fraction of the wavelength). Because there is considerable overlap in the scales, it may be possible to read the antenna capacitance (and losses) at two different frequencies. In that case, choose the reading that is closest to the desired operating frequency. You can make an estimate of the losses in the antenna by noting the resistance reading at the resonant point. This resistance includes the series loss resistance of the reference inductor inside the analyzer. To find the inductor's loss resistance, disconnect the antenna, and connect a capacitor that has approximately the same value as the antenna capacitance to the analyzer's output terminals. Now when you tune for resonance you will be reading the loss in the inductor. The external capacitor will also contribute some resistance, but if it's a low-loss type (such as an air variable, silver mica or polystyrene), the measurement error will be less than 10 per cent. Subtracting the loss resistance of the inductor from the measurement obtained on the antenna will give the losses in the antenna alone.

Except for some types of disk ceramics, most small capacitors have very low series resistances, and their losses can't be measured accurately with this analyzer. The series resistance of the internal reference inductor usually will be much larger. But the analyzer is still useful for testing unknown capacitors from your junkbox. If you're suspicious about the loss of a capacitor, use the technique described above for determining loss in an antenna.

When measuring inductors, especially loop antennas and large-diameter loading coils, keep the inductor away from objects that will disturb either the magnetic or electric fields around the inductor. The biggest effect will come from large metal surfaces, but even rather poor conductors like body parts of experimenters will affect the loss measurement if they are too close to the inductor. By the way, the same admonitions about keeping inductors away from lossy surfaces apply when mounting the loading coil on your antenna. To measure the inductance and series loss resistance of a coil, set S2 to the appropriate position. If you don't have a good advance estimate of the inductance, you will have to try each of the "L" positions and sweep through the frequency range to find the null. Even though the series loss resistance of an inductor is actually what we need for circuit calculations, most experimenters are used to thinking in terms of coil Q. Once you know the inductance and the series loss resistance RL of the coil, the Q is calculated from: Q = (6.28 x F x L) / RL, where F is the frequency in kHz and L is the inductance in millihenries. Both the Q and loss resistance will vary with frequency. You can determine the series resistance of an inductor at a specific frequency if you have a low-loss air-variable capacitor (or a good assortment of fixed low-loss capacitors). Set the dial of the analyzer to the desired frequency with S2 in the "DIRECT" position, connect the inductor and the air-variable capacitor in series, and tune the variable capacitor for the minimum meter reading.

Inductors which use ferrite or powdered-iron cores may exhibit some core saturation effects even at the low power levels used in the analyzer. This can result in a loss resistance measurement which is higher than what the inductor would actually have at very small signal levels, such as in a receiver. By the same token, the loss resistance of a ferrite-core inductor used as an antenna loading coil may be much higher even at 1-watt power levels than the value you measure with this analyzer or another instrument.

Alternate VFO Circuit -- I built the first prototype of the LF analyzer with a conventional Hartley oscillator. It provides a stable and constant sine-wave output over a supply voltage range of 6 to 15 volts, and draws less current than the "sine/triangle/squarewave" generator. However, it has become very difficult to find an assured supply of 365 pF (or larger) variable capacitors. For experimenters who have their own stash of variable capacitors, the Hartley oscillator circuit is included here as Figure 4. This circuit can be used in place of U1A, U2, U3, Q1 and their associated components. The 100-ohm resistor and 0.1 uF capacitor in the emitter-follower output circuit correspond to R10 and C8 in Figure 2. L1 is a Mouser number 434-06-472J 4.7 mH inductor, and L2 is 20 turns of # 30 wire wound on the body of L1. If the oscillator won't take off, try reversing the leads on L2.

SWR option -- Unlike hams, who are often concerned (some people might say obsessed) with standing-wave ratio, LowFERs have much less need for that type of measurement. SWR is important when you have long transmission lines, which don't make sense on the LowFER band because transmission-line length has to be subtracted from the allowed antenna length. Also, most solid-state ham transceivers are designed to work into a 50-ohm load and have built-in protection circuits that reduce the RF output when the SWR is too high. You can get full output only if the impedance seen by the transmitter is fairly close to 50 ohms resistive. With a properly- designed LowFER transmitter that won't put out more than 1 watt, there's not much concern about burning out the finals. The antenna does not have to present any special impedance to the transmitter. All that matters is that the load seen by the transmitter is resistive rather than reactive, and that the transmitter doesn't have to deliver higher voltages or currents than the final amplifier can handle. The complementary pair final that I use on LF will operate into resistive loads from below 15 ohms to over 100 ohms without much change in efficiency. So, although I like to know the resistance of my antenna circuit, I don't really think in terms of SWR. For those who are interested in SWR measurements at LF, the broadband resistance bridge circuit of Figure 5 can be used with the signal generator and meter amplifier portions of the LF analyzer circuit. Figure 5 is a variation of a circuit that appears in the ARRL Antenna Book. A null in the "reflected" voltage occurs when Rb/Ra = Rx/Rs. With Ra = Rb, the DC voltage at the "reflected" output will be zero when Rx = Rs. To use this circuit as a typical 50-ohm SWR bridge, Ra and Rb can be 330 ohms, with Rs = 50 ohms.

Using a variable resistor with a calibrated dial in place of a fixed resistor for Rs increases the versatility of the circuit. It is also possible to use unequal resistors for Ra and Rb to provide a convenient multiplying factor so that easily available potentiometer values can give any desired measurement range. For example, if Ra is made equal to twice the value of Rb, the null in "reflected" voltage will occur when Rx = Rs/2. With Ra = 300 ohms, Rb = 150 ohms, and with a 1K audio taper pot for Rs, the effective "characteristic impedance" for the measurement can be varied from zero to 500 ohms. In this case, the Rs dial can be calibrated using an ohmmeter across Rs (with the other components disconnected) and dividing the resistance readings by two. By alternately tuning the generator frequency and Rs for a complete null in the reflected reading, you can read the RF resistance of the antenna at resonance from the Rs dial. This provides an alternative to using a calibrated meter dial to read the resistance, and is independent of the meter gain setting. A complete null will occur only if the unknown circuit connected to the bridge output terminals is purely resistive at the measurement frequency. A 0.02 uF capacitor connected to a 50-ohm SWR bridge and measured at 159 kHz will not produce a null in the reflected power, even though it has a 50-ohm impedance at that frequency. Because a purely capacitive or inductive load does not dissipate any power, it should give an "infinite" SWR reading.

To make conventional SWR measurements, set Rs to the value corresponding to the characteristic impedance (such as 50 ohms) that you want to use for your SWR measurements. Connect the antenna to the output and tune for a minimum meter reading with the switch in the "reflected" position. Then switch to "forward" and adjust the meter gain for full-scale deflection. Now you switch back to "reflected" and read the actual SWR from the meter. SWR calibration is accomplished by connecting known resistors across the output. For example, when the characteristic impedance of the bridge is set to 50 ohms, the meter should indicate an SWR of 2:1 with loads of either 25 or 100 ohms, 1.5:1 with loads of 75 or 37.5 ohms, etc.

Accuracy -- This analyzer is not nearly as accurate as a laboratory bridge circuit or a digital L-C meter such as the one reviewed in the April 1997 LOWDOWN. However, it makes up for this by allowing you to make inductance, capacitance and loss measurements at or near the actual operating frequency. The frequency accuracy at room temperature is as good as your calibration. There is likely to be some frequency drift with temperature, and a spot check of frequency with a receiver or counter may be advisable if the analyzer is used outdoors at very low or high temperatures. Inductance and capacitance measurement accuracy depends strongly on the frequency accuracy. A 2.5 per cent error in frequency will result in a 5 per cent error in the inductance or capacitance reading. Assuming correct frequency calibration, the inductance and capacitance accuracy also depends on the tolerance of the internal reference capacitors or inductors, which are rated at 5 per cent. As with many instruments, it is necessary to subtract a constant offset from the measured value of capacitors, to account for stray capacitance in the measuring circuit. Distributed capacitance in the reference inductor causes an additional offset in capacitance measurements. Subtracting 20 pF from the capacitance values indicated on the dial should allow an overall capacitance measurement accuracy of better than 10 per cent. Spot checks on known capacitors will allow you to determine the necessary corrections if better accuracy is required.

RF resistance measurement accuracy can be improved by using an external meter with a larger scale. Accurate measurements of the series resistance of high-Q circuits are difficult because of problems mentioned earlier with harmonics in the generator output, which affect the depth of the null reading when the circuit is tuned through resonance. You should also be aware that the meter and detector circuit are calibrated only for resistance, and will not give an accurate reading of the impedance of a capacitor or inductor. However, the analyzer can probably provide better than 10 per cent accuracy when reading the resistance of a series L-C circuit with a Q less than 300 (such as a typical LowFER antenna) at resonance. For best accuracy, use the internal reference resistors to recheck the meter calibration at the frequency of interest. First make the measurement on the unknown component or circuit to find the resonant frequency and to get an approximate reading of loss resistance. Then short the output terminals and adjust the meter gain at that frequency, using the internal reference resistor that's closest to the value you measured. After that, you can go back and recheck the meter reading on the actual circuit.

Because of distributed capacitance effects, an inductor will typically show a higher value of inductance when measured on this analyzer than it would on an instrument that operates at a lower frequency. This isn't necessarily an error -- the effective inductance of a coil actually does increase with frequency. Computer programs are available which will give accurate predictions of coil inductance versus frequency. The program I use is COIL.EXE by K6STI, and I believe it is now part of the software package that comes with the ARRL Handbook. COIL.EXE predicts that an air-wound coil similar to one I use on LowFER beacon LEK has an inductance of 2.5 mH at 1 kHz. It rises only to 2.56 mH at 100 kHz, but is 2.75 mH at 200 kHz (a 10 per cent change) and 3.14 mH at 300 kHz (a 26 per cent change). Even though it's not a precision instrument, this analyzer may actually provide a more meaningful reading of LowFER loading coil or loop antenna inductance than a 1-kHz laboratory bridge.

Extending the frequency range -- The sine/triangle/square wave generator used in the analyzer circuit will not work above a few hundred kilohertz because of speed limitations in U1 and U2. However, the generator and the rest of the circuit are capable of operation all the way down to audio frequencies. To extend the frequency range to cover the UK 73 kHz band, C6 can be increased to 680 pF. For measurements in that range, I also recommend adding a 22 mH reference inductor and perhaps eliminating the 680 uH inductor. Because antenna system losses are likely to be fairly high at 73 kHz (primarily due to increased losses in the loading coil), it may be advisable to increase the value of R10 to 200 ohms and to calibrate the meter for resistances up to 300 ohms.

Parts list -- The list below includes all of the parts needed for the project except for knobs, which can be obtained from any of the three suggested vendors, and the mounting screws for the PC board. Total cost of the components in the parts list, excluding tax and shipping, is approximately 30 dollars, based on current catalog prices as of April 1997.

Vendor codes used in the parts list are as follows:
R = Radio Shack
M = Mouser Electronics (1-800-346-6873; http://www.mouser.com)
D = Dan's Small Parts (Box 3634, Missoula, Montana 59806-3634; Phone or Fax 406 258 2782; http://www.fix.net/dans.html)

LF Analyzer Parts List:

Schematic Ref.  Qty  Vendor  Part No.         Description

Q1               1     M     333-PN2222A      2N2222A Transistor
T1               1     M     42TL030          100 -100 ohm transformer
D3, D4           2     M     592-1N4148       1N4148
D1, D2,
  D5, D6         1     R     276-1123         1N34A (package of 10)
C1, C4           2     M     539-TKR50V10     10 uF 25 volt capacitor
C10              1     M     5989-100V1.0     1 uF capacitor
C11              1     M     23PS122          220 pF polystyrene capacitor
C12              1     M     23PS168          680 pF polystyrene capacitor
C13              1     M     23PS222          2200 pF polystyrene capacitor
C2               1     M     539-TKR16V100    100 uF 16 volt capacitor
C3, C8, C9       3     M     581-EXWD104M     0.1 uF capacitor
C5               1     M     141-100N2-015J   15 pF capacitor
C6               1     M     23PS133          330 pF polystyrene capacitor
C6a              1     M     242-8050         8-50 pF trimmer capacitor
C7               1     M     140-CD50S2-033J  33 pF capacitor
L1               1     M     434-17-682J      6800 uH choke
L2               1     M     434-17-222J      2200 uH choke
L3               1     M     434-17-681J      680 uH choke
R14, S1          1     M     31CC501          100K pot w/switch
R5               1     M     31VA401          10K linear pot
R6               1     M     569-72XL-10K     10K trimmer pot
S2               1     M     10YX112          12 pos switch
U1               1     M     570-CA3240E      CA3240 dual op amp
U2               1     M     511-LM311N       LM311 comparator
U3               1     M     513-NJM2930L05   2930L05 low-dropout 5V regulator
R18-R22          5     M     271-xxx          1/4 W 1% resistors (xxx = value)
R1-R4, R7-R13,
  R15-R17       14     M     29SJ250-xxx      1/4 W 5% resistors (xxx = value)

Other:           1     R     270-627          Molded case
                 1     R     274-662          Binding posts (package of 2)
                 1     R     276-150          Prototype board
                 1     D                      200 uA edge-reading panel meter
                       M     391-0301         alternate panel meter (500 uA)
                 2     M     571-26404633     8 pin DIP socket
                 1     M     123-6004         9 V battery clip
                 2     M     34-425           3/4 inch # 6 threaded spacers

(The parts list is also available in graphical form.)

(This article originally appeared in The LOWDOWN, and is protected under U.S. and international copyright laws. Obtain the author's permission before distributing in any form.)

Return to Part 15 Library.
Or, to the Longwave Home Page.