Skip to content

Variations on a Dipole

August 5, 2017

Here are the slides from a presentation I gave to the Valley Radio Club on August 4, 2017.

Variations on a Dipole – Power Point

Variations on a Dipole – PDF

Antenna patterns are all about interference

Modifications to the IC-745 for the Digital Modes

July 29, 2017

The IC-745, like the older ICOMs, is built very robustly with parts that you can see without a magnifying glass, and which you can check with standard meters.  The schematics are available on-line, the old ICOMs come up for sale on e-bay regularly, and there are support groups on line to help out when problems arise.  The major problems are know and the fixes are documented.  Although the IC-751 has the advantage of serial computer control that the IC-745 does not have, the IC-745 can also be a very good first HF radio for those willing to do a little restoration.  These radios made in the early 1980’s are some of the first all solid state transceivers and represented state of the art technology when they came out.  To this day, they perform well for the functions that they were designed for, mainly SSB and CW communications.  However, the ham world changed in the last thirty years with the advent of the PC and the digital modes.

There are a few things that you need to accomplish to have the best possible operation on digital modes with the old radios.

  1. Computer control of the radio (CAT).  Sorry, we have to give that one up on the IC-745.  It just was not made with serial control in mind.  Arduino freaks could conceivably make that work – but you would be taking on a giant project.
  2. Sound Card connections.  Getting audio to and from the transceiver is required for digital modes.
  3. Ensure stability and true frequency calibration.  The narrow band digital modes care more about frequency shifts and drifts of a few Hertz which voice communications would never have noticed.
  4. Extras like an RTL-SDR panadapter  can provide modern functions to vintage radios.

Having given up on #1 lets get on to the others.

Sound Card Connection for the IC-745

Schematic section showing the ACC connector

To work the digital modes you need to be able to connect signals to and from a computer sound card.  It is always possible to use the MIC connector, but then you will be always swapping out this connection when going between voice and digital modes.  Better is to use the rear 24 pin ACCessory connector.   The audio output from the IC-745 can be found on Pin-4 of the ACC connector.  Ground is on Pin-8 next door.  I directly wire an audio cable from Pin-4 to the sound card Line IN.  It is a little harder to figure out where to put the sound card Line OUT.  This transceiver was never designed to have anything other than a mic input, so we will need to play tricks.  The MODE wire, Pin-5, is close to what we want.  If you follow it, it comes from the one side of the MIC gain potentiometer.   That seems promising until you realize that this point is the output of the mic preamp (wire MIC1).  My solution was to sum the mic preamp and the MODE wire together through resistors at the high side of the mic gain pot.  There is already a resistor on the MODE wire, R75, but  none on the MIC1 wire.

MIC and VOX amps. MIC2 wire is modulation input after MIC gain control.


MIC Gain pot

If you stare at the bits of schematic I’ve provided, you can figure out this logic.  I added a 1.5kΩ summing resistor to the MIC1 wire at the J8 header.  This allows the MODE wire to be an audio input controlled by the MIC Gain knob yet still allows signals from the MIC preamp to also be used.  The photo below shows how this was done on the J8 header.

With this simple modification, the ACC connector Pin-5 will work as the audio input form the sound card Line OUT. The VOX will also work using the sound card audio to key the rig. The MIC connector is free for the mic. Just be aware that the ICOM mic is always live, so if you do leave it plugged in, keep quite when transmitting via the PC sound card.

Resistor added to MIC1 wire on J8

Improving Frequency Stability on the IC-745

These early ICOM radios use a phase locked loop (PLL) stabilized VFO, which was state of the art when they were made.  Unfortunately, the PLL circuits used some plastic trim capacitors which are guaranteed to deteriorate over time.  They need to be changed out if they haven’t been already.  A very good description of the trimmer replacement and other useful mods and adjustments written by Frank of W3UHF can be found here.  Even with the trimmers replaced, the IC-745 suffers a small problem caused by thermal changes near the crystals that control the master oscillators in the radio.  The problem is worse with radios with the internal AC power supply.  The issue is that when the fan comes on to cool the output transistors, air is pulled in the vents in the bottom of the enclosure and the cool air flows over the crystals.  They change frequency just a little.  When the fans stops, the crystals warm up again and the frequency drifts back the other way.  I found that isolating the crystals from the air flow with some Styrofoam “peanuts” help quite a bit.  There are two crystals to protect for the 1st and 2nd local oscillators.  I still see a little general temperature drift, but no longer on the ~1 minute time scale that made it difficult to hold a digital QSO.

Foam peanuts wedged over the LO crystals

The architecture of this radio makes it difficult for the TX and RX frequencies to be spot on.  This is mainly an issue on the PSK31 mode where being off by 10Hz will mean you may have difficulty making a contact.  I did the best I could with the LO frequency adjustments, but finally just resorted to using the RIT control to get things perfect.

RTL-SDR Panadapter pick-off point

1st IF Mixer and Filters. Pick-off point across R20.

RTL-SDR dongle pick-off

The 70 MHz first IF frequency on the IC-745 provides the perfect place to grab a signal for the ubiquitous RTL-SDR dongles that are available.  With the RTL-SDR you can quickly see all of the signals on the entire ham band to which you are tuned.  A little study of the schematic shows the appropriate pick-off point.  You want to be right after the 70 MHz mixer but before the IF filters so that the dongle can see a wide bandwidth. In the schematic fragment, the point I chose was across resistor R20.  I didn’t want to disrupt the circuit, yet need to get signal into a piece of co-ax, so opted for a 100 pf and 1.5kΩ series coupling into the 50Ω co-ax.  I did not see any deterioration of the IC-745 signal doing this.  In the photo you can see capacitor and resistor tacked in.  The sub miniature coax, since moved over to my IC-751A, found ground on the bent over leg of R20.

Go forth and have fun!

Modifications to the IC-751A for the Digital Modes

March 6, 2017

The 30 year old ICOM HF radios still have a bit of life in them if given the chance.  I’ve managed to work all of the states and more than 140 countries using the digital modes with my old IC-751A in the last couple of years.  A few small modifications have helped to run digital modes with this radio.  This post will describe in detail the physical modifications and additions that I’ve done to the IC-751A to use it with a computer running FLDIGI or similar digital-mode programs.  These mods should work for the IC-751 as well.

Sound Card Audio Connection: Without modifying the radio, you have two choices.  You can use the rear panel auxiliary connector, Pin-5 for audio input, but you get no front panel knob to control the input level, nor can you use the VOX for PTT control. Or you can use the mic connector, but then you will always be disconnecting / reconnecting when you want to switch between the microphone for phone and the computer for digital modes.

Aux audio connection rerouted to VOX pot.

Two-pin connector for audio re-route

For these reasons, I made a small modification to the 751A wiring that allows use of the rear connector, but plumbs the audio input signal through the MIC GAIN control and VOX circuits.  The pictures show the plan.  The wire from Aux Pin-5, the trace labeled MO, is lifted from connector J5 and attached to trace VOXG on J7.  I did this with the aid of a two-pin connector so I could easily back up if I needed to.  As the picture shows, I picked off the ground from the shielded but ungrounded VOXG cable and also tapped off the connection at J7 Pin 1 for the two-pin connector for the MO wire. The last picture shows the MO wire rerouted to the two pin connector, ready to be covered up with shrink tube.

Wire with red sleeve is MO wire from AUX pin 5

Wiring re-routed

Wired this way, both the VOX GAIN and the MIC GAIN controls will affect the level of the audio input signal.  Using the VOX is a handy way to key the rig when transmitting using the digital modes, so you should turn the VOX GAIN up enough that the digital audio signals from the PC reliably trip the VOX and key the transmitter.  Then you can use the MIC GAIN control to fine tune the audio level into the transmitter.  For many digital modes it is important to keep the RF amplifier linear and not overdrive it.  I achieve this by leaving the RF PWR level control turned up all the way while using the MIC level control to then set the power output.  If the rig wants to throttle your output, it will do so by raising the ALC level.  This can happen if the SWR match is poor, or if you try to drive the output beyond the capability of the rig or beyond the RF power setting.  Ideally, the ALC level reading on the meter should never lift off of zero for clean digital transmissions.  But this means watching your output power level and using the MIC GAIN knob to adjust the input level, rather than throttling with the RF PWR control.

To complete the connections to your sound card, you will need to wire up the rear panel 24 pin auxiliary connector.  You should connect “Line-IN” of the sound card to the AF OUT on AUX pin 4 and the “Line-OUT” on the sound card to the MO line on AUX Pin 5, using the common ground, AUX Pin 8, for both signals.

Serial Interface:  If your radio doesn’t have one, get one!  The good news is that Piexx makes a replacement for the long discontinued ICOM UX-14 option that is better than the original.

PTT from Piexx board

Piexx UX-14px board with PTT wire (purple) routed to top side of chassis

The new Piexx boards provide CAT controlled PTT and support reading the S-Meter.  I’ve modified the FLDIGI rigcat IC-751.xml file to take advantage of these capabilities.  If you wish to use the Piexx board’s PTT function, you can easily tap into the rig’s PTT line.  The PTT line is called SEND on the schematic and can be found at J8 Pins 4 & 5.  (You can find J8 on the schematic fragment shown in the picture above.)  The pictures show the connection to the J8 header and wire routing.  The S-meter connection is good to install as well, especially if you ever contemplate running your old rig remotely, which you could conceivably do with the serial connection and a remote desktop application.

RTL-SDR Dongle Connection:  To add some real additional functionality to this old rig, attaching a $25 RTL-SDR dongle on the IF has to be the biggest bang for the buck that you can get.

RF section schematic showing SCOP connection

The SCOP connector J4 on the RF board

Coax routed from RF box and through the bottom cover

It is easy to do with the IC-751A.  Just find the SCOP connector point and attach a piece of coax to go into the antenna port of the dongle.  The pictures tell the story.  There is a nice hole in the bottom cover that lets you get the cable out.


You are ready to have fun!



Fast Avalanche Transistor Pulser

December 18, 2016

You never know when you might need to measure the speed of light.  When the time comes, this avalanche transistor pulse circuit will come in real handy.  This is one of my favorite little bits of electronics because you get such great performance out of a single common 5¢ transistor that is abused for this purpose!  Many bipolar transistors will exhibit avalanche breakdown if the voltage gets high enough on the collector, but you won’t find many that give a much better pulse than the common 2n3904. pulser_sch


Avalanch transistor waveform with ~2 ns FWHM


Timing pulse returned from open end of a 20 ft. length of coaxial cable.

The schematic shows the circuit.  I built my version of this pulser into an existing timing box I had made before.  You need a modest DC voltage to make the transistor avalanche so I used one of the primary windings on a small AC line transformer as the source for the ~150V DC supply.  The charging resistor, R1, limits the current to the transistor.  Above about 115V the 2n3904 will start to conduct, but will be quite happy as long as the current is limited by R1 to a few tens μA.  A small 1:1 ferrite core transformer of a few turns works well for triggering the transistor.  The base-emitter junction is turned off hard by the DC short of the transformer.  When triggered, the transistor undergoes an avalanche breakdown with < 1 ns rise-time, discharging C2 through the emitter resistor load.  You can trigger at >1kHz rep-rate without the pulse losing amplitude.  I arrange the load to tap off a fraction of the discharge voltage and also to appear as a matched 50Ω load to absorb any reflected pulse returning on the cable.  The discharge capacitor, C2, can either be a small, few pF cap, or it can be a short piece of transmission line if you want a square pulse. I like the spike, so went with 22 pF into about 50Ω which discharges the capacitor in about 1 ns when the transistor switches.  The waveform I see on my scope with 400 MHz bandwidth is a smoothed pulse about 2 ns wide into a 50Ω load. If I had a faster scope it would be even sharper.

What good is it?  Well, if you are after a low-jitter fast-rising edge trigger, you can’t beat that spike.  The pulse is also very handy to time cables, to adjust them for identical or specific electrical lengths, or just to find out how long a piece of cable actually is without getting out a tape measure.  The second oscilloscope trace shows the unmistakable reflection from the open end of a length of coaxial cable.  The measurement functions on the scope says that it took 61.2 ns for the pulse to traverse the cable and return.  RG-58 cable with polyethylene dielectric has a velocity factor 0.66 the speed of light.   Hence the length of cable can be calculated as:  L = 61.2 ns / 2  *  3.0 x 10 * 0.66 = 6.06 m = 19.9 ft. which is pretty close to the length I measured at 19 ft. 8 in.

If you get out your nippers, you will find that you can trim cable lengths to within just a couple of inches of the length you want without need of a tape measure.  I find it pretty amazing that we can tell how long it takes light to travel over just a couple of inches of wire…  Along with a decent oscilloscope, a fast pulser makes all the difference for this kind of measurement.

Using the RTL-SDR Pan-adapter

September 16, 2016

I have had a chance to use HDSDR a little more with the RTL-SDR pan-adapter for my IC-751A transceiver and have found it a very nice tool for a number of applications.  Here are a few examples.

Pan-adapter for rapid SSB tuning

This is the obvious application.  Using the RTL pan-adapter with HDSDR set to “Full sync in both directions,” it is possible to monitor the entire ham band for SSB signals and to tune the parent radio to the transmission by clicking on the signal in HDSDR window.  The received signal can be heard either through the parent radio or demodulated with HDSDR.  Either the RTL-SDR or the conventional receiver might sound better, depending on the quality of the parent receiver.  The filter edges are very steep and simple to adjust in HDSDR, and there is a very sharp notch filter that can remove unwanted narrow-band signals that might be contaminating the voice channel.  Nevertheless, to my ears the wider audio notch in the 751A seems to provide more readable and comfortable voice signals.


The HDSDR screen shot above shows more than a dozen voice signals to sample across the 20 meter band.  One of the nice features of this program is the easy way to set the squelch with a click on the S-meter. A curious digital mode transmission is  in progress just above the ham band at 14.366 MHz.  For the voice modes, the frequency synchronization and stability between the parent receiver and the RTL-SDR radio is plenty good enough that your transmitted signal will fall on top of the HDSDR frequency settings (once you have it all set up correctly) if you are listening using HDSDR.

RFI identification and signatures

The big view of the pan-adapter can show more than just intentional transmissions. In the screen shot below you can see the situation on 30m this afternoon.  The wide view makes RFI problems unequivocally show their presence.  Below you see a strange on/off modulated RFI signature every ~17kHz across the band.  RFI can come in many forms.  Switching power supplies can generate  bands of noise spaced at uniform frequency intervals.  One of the next projects here will be to set up a portable RTL-SDR computer rig for chasing down such RFI sources, which can plague reception especially on the lower bands in my neighborhood.


Working the pile-up

To the left of the center of the waterfall above you can see a pack of CW operators in a “pile-up”, all looking to contact a rare DX operator.  In this case a “DX expedition” to the South Sandwich Islands is the attraction.  Breaking through the pile-up is an art that I have yet to master, but I can see that the wide view of the pan-adapter can be invaluable to help identify the actual station of interest in the midst of the hoards of other signals seeking attention, as well as the frequency where the operator is listening by identifying where in the pack the selected stations are replying.  If the DX station is working “split,” outside the 3kHz bandwidth of the 751A, picking the correct place to call can be purely guess-work without the wide-band knowledge.

Monitor SSB net while working PSK on the parent radio

The are many examples of “doing two things at once” that can be accomplished because of the ability of HDSDR to be tuned independently of the parent receiver. For starters, tune the transceiver to the PSK portion of the band.  Select independent tuning for HDSDR.  Configure the audio output from HDSDR to the computer speakers and go hunting on the voice section of the band.  Listen in on your friends while you use FLDIGI to work the PSK section of the band with the parent transceiver.

Monitor the JT65 & JT9 band while working PSK – or vise versa

It is possible to run two digital-mode decoding programs, one for the parent transceiver and one for the RTL-SDR receiver tuned through HDSDR.   Yes, you will have three radio programs going at once!  Sometimes it is not possible to get everyone to CAT correctly – so here is my general plan.  The working application will be connected to the transceiver audio in the conventional manner, and will be controlled from HDSDR using the “CAT to HDSDR” port via a Virtual Serial Port (VSP).  The monitor application will get it’s audio from HDSDR via a Virtual Audio Cable.  You will have to manually tune this monitor application to match the HDSDR tuning since the working applications is using the “CAT to HDSDR” port.  Use “independent tuning” so that the HDSDR’s LO control tunes the transceiver and the working application, while HDSDR’s Tune control tunes the monitor application.  Below are step-by-step instructions for setting this up:

  1. Configure HDSDR and Omni-Rig to CAT to the transceiver.
  2. Set up a Virtual Serial Port between “CAT to HDSDR” and working program you will be using to work contacts, e.g. FLDIGI.
  3. Set that program up to accept VSP connection as a Kenwood TS-50 as HDSDR requires.
  4. Test that you can control the rig and HDSDR via the working program, FLDIGI for example.  Changes from the rig will reflect back to HDSDR which will then make the changes to FLDIGI.  Similarly going the other way, HDSDR is in the middle.
  5. Set HDSDR sync mode to “independent tune in HDSDR.”  Note that tuning the LO frequency control will change the transceiver frequency.
  6. Bring up the monitor application, e,g. WSJT-X.  Set up for no CAT control.
  7. Connect the monitor application to HDSDR audio output using a Virtual Audio Cable.
  8. Tune HDSDR to the band section you wish to monitor.  Manually tune the monitor application to the same frequency.
  9. Go back to the working application and make some contacts.  You can tune around the band with the working application all the while the monitor application remains at the desired frequency.  However, if you change bands you will have to re-tune the monitor application.

I almost always try to work stations with audio routed to/from the transceiver in the conventional manner.  It is possible to listen to the RTL-SDR / HDSDR demodulated audio, and then transmit using the parent transceiver.  However, my RTL-SDR dongles do not have the frequency stability that allows me to know for sure that I will be exactly on-frequency.  A frequency shift of just a few Hz between Rx and Tx can lead to confusion with digital modes.  Using “split” modes can help deal with Rx/Tx discrepancies.

Leave the radios on and monitor PSK, RTTY, JT65 & JT9 traffic for PSK Reporter

You can use the set-up above as a much more complete digital-mode band monitor.  Set the applications, e.g. FLDIGI and WSJT-X to report spots to the PSK Reporter website.  JT65 and JT9 spots are the most productive, but there are plenty of PSK signals as well that might be of interest that FLDIGI can flag.  Once your spots are logged to PSK Reporter, you can quickly see what parts of the world you are hearing via the PSK Reporter website map.

I’ve also used RCKskimmer, to monitor the band for RTTY and PSK31 & PSK63 signals.  If RCKskimmer is looking at the RTL-SDR radio tuned through HDSDR, then you can skim signals from the PSK and RTTY sections of the band.  Meanwhile, the transceiver can be tuned on the JT65 region.  Again, all spots can be sent to PSK Reporter.

Using these techniques you might be amazed at what your antenna is picking up when you otherwise are not actively listening.  I’ve started using WSJT-X 1.70 recently which has a much better decoder than previously. I’ve seen more than two dozen signals simultaneously decoded when looking at the full JT band with the RTL-SDR using the new WSJT-X.  The screen shot below shows typical activity seen on 20m over a few hours monitoring both the entire JT65 & JT9 bands and the PSK band.  If I keep the radio on I reliably end up in the top 20 monitors in the PSK Reporter statistics despite my out-of-the-way location in the Pacific Northwest.


Mapping the Radiation Pattern of a Fixed Long Wire Antenna

June 14, 2016

The simple 20 meter pull-up vertical antenna I described last time has been working well.  I’ve been putting my efforts into working DX stations and have found that the vertical often does almost as well as the Off Center Fed Dipole (OCFD) despite its generally lower gain.  Part of the reason is that there are places the OCFD does not hear very well, such as along its pointing direction, north into Russia.  Also, the OCFD seems to do especially well to the east coast and southeast, and at times the band can be overwhelmed with signals that I am not interested in.  With these effects taken together, the vertical dipole can sometimes offer better performance than the generally higher gain dipole.

Part of the motivation for the vertical antenna was so I could better map my OCFD antenna.  When I tried to do that exercise comparing to my 40 m loop, there was too much uncertainty in both the loop model and the OCFD model to be able to make much sense of the results.  However, comparisons with an inherently symmetric vertical antenna can directly generate the radiation pattern for the horizontal wire.  So that was the plan; collect lots of JT65 and JT9 signals simultaneously on two receivers connected to the two antennas and make comparisons for lots of geographic locations.

Needless to say, it takes a certain amount of persistence to churn through the data collected over a week or two of leaving the radios on.  Please see the previous post and the  Excel template if you want to try this yourself.  All together more than 67,000 signals were recorded.  More than 15,000 acceptable simultaneous signal report pairs on the two antennas from more than 450 unique grid squares went into the mix to generate the plot below, where report pairs for the same four-character grid square were all averaged together.

OCFD Pattern

Antenna radiation pattern generated by comparing identical transmissions from many independent transmitters using a vertical dipole antenna as the uniform reference antenna.

It is possible to identify several strong lobes and nulls which are expected when harmonics are generated on long wire antennas.  But it is not the way my OCFD was supposed to look!  I wanted the strongest lobe at about 30° to look into Europe, but instead the best direction is toward Florida.

When I put up my wire, it got revised a few times along the way and I never pulled the entire antenna down to carefully measure the length again.  Hence, it is certainly possible that I measure or cut something wrong, or that there is a systematic problem with the electrical length of coated conductors so that the antenna up in the trees was different from what I intended.  There are also many details not included in the model, such as rain gutters, other antennas, and ground variations.

Rather than despair, I thought I would see if I could generate a similar pattern with the NEC model of my antenna if I just changed the lengths of the radiators by a small amount.   So I ran the NEC simulations and lo and behold, the patterns generated when the antenna is too short about 3 feet are really pretty close to what I am observing.

OCFD Pattern vs Tail L

Radiation pattern for OCFD when length of short section of dipole is changed in 1 foot intervals.

The red curve, with 37′ tail section was the desired design.  Comparing the strength of the lobes at 105, 80 and 30 degrees with the actual data in the first graph, you can see that the measured radiation pattern seems to reflect the brown curve the best – which is modeled with a 34′ tail section.  You get similar results if you change the length of the long section instead of the tail section.

It was a simple matter to add another four feet of wire to the antenna tail.  That should put me on the blue curve which would significantly improve reception into Europe and significantly degrade reception to the east coast of the US.  The Pizza plot for my location with the two overlaid antenna patterns tells the story.  I’m giving up larger side lobes for the “cross” lobes.  The northern “cross” lobes  enjoy a considerable enhancement from what they were.


Antenna patterns overlayed on polar map. Adding three feet to the antenna produces the pattern with the exaggerated “cross” lobes.

Needless to say the radios are on again, listening to JT65 and JT9 signals to see if I’ve really understood this well enough to get improved performance into Europe.  Assuming I’ve now changed the antenna pattern, I will miss the strong signals from the southeast US, the Caribbean islands and Australia (Tasmania is now in a deep null).  However, I will be leaving the vertical dipole up in the trees so with a flip of the switch I’ve pretty good general coverage.

When we string up a wire antenna, we often just hope for the best without any way of making a measurement of the antenna performance.  The methods described here shed light on what was before shrouded in uncertainty, and gives enough precision that we can use the method to iteratively improve our antenna installations.

UPDATE  7/2/2016

I have analyzed the data from the modified antenna and I did indeed make some improvement to the strength of the northern lobe into Europe.  Below is a comparison of the data for the two runs.

Both Patterns

Although the northern performance did improve, the pattern looks less like the model than before.  Particularly, the null at 90 degrees got deeper – not expected in the model runs, and the lobe at 105 degrees is still pretty strong.  Maybe I didn’t go quite far enough with the change?  But definitely there is some improvement, and the ~3-4 dB increased gain northward has helped considerably with contacts into Europe.

Although not perfect, the ability to map a wire antenna, compare with NEC models and then make small modifications can actually lead to significant performance enhancement.


A Simple Pull-up Mono-Band Verticle Dipole Antenna Design – and 4NEC2 Tutorial

April 21, 2016

Vertical dipole with transmission line feed and choked feed line. RF current magnitude in green.

I have a nice off-center-fed dipole (OCFD) antenna that is my mainstay for multi-band HF work, but I would like to understand it better.  With a wire antenna tied to trees it is not a simple matter to quantitatively determine the antenna’s radiation pattern experimentally.  One way to get a handle on this is to compare signal levels against an isotropic antenna.  Vertical antennas are nice this way, since you just need a wire straight up in the air and you naturally get an azimuthally symmetric pattern.  This was my motivation to build a vertical dipole.

When you look around at how people put up vertical antennas, most of the talk is about radials – how many, how long, raised versus buried  – so it goes.  My gut feeling is that all this effort is misplaced.  The problem with vertically polarized radiation is that when the EM wave impinges on ground it induces “radial” currents.  If the antenna is close to the ground we can provide radial wires so the RF currents can return, a counterpoise for the currents in the  1/4 wave vertical section.   I have tall trees.  This means I can just pull the vertical antenna up into the trees so that the EM wave isn’t centered on the ground plane.  I’d rather not launch a wave into the ground anyway — I want the wave to propagate across an ocean or two!  Traditional verticals with  radials will have the same problem with the EM wave propagating near the ground surface once the wave has moved past then ends of the radials, so why spend all of this effort on connecting the antenna into the ground for a few meters when you can’t do it for the next 1000 km anyway?!    As soon as I pull the antenna up into the trees by at least 1/4 wavelength, I can now have a vertical dipole if I drop a wire from the feed point back toward ground.  But rather than using another wire, realize that I already have a wire coming back down from the feed point, namely the outside of the co-ax feed itself.  The important thing to do is to define the end of that radiating segment for RF, so there we place a choke.  That’s it– our feed-line coax with a choke about 1/4 wavelength from the feed point and another 1/4 wavelength wire continuing on up — pretty easy!

With 4NEC2 I model because I can.  This simple antenna system is a great sandbox for learning the tricks of the NEC codes and some of the basic principles of antenna design, and to find all the pitfalls of this nice idea.  My NEC description for this vertical antenna model is located here, to use with 4NEC2 available here.  I’ve included some “extras” like transmission lines, chokes, and wire coating in addition to the antenna wires.

When I write the model definition I like to use variables rather than just specify fixed numbers.  I try to express the dimensions as functions of things I might like to change, for example the height of the top of the antenna.  I’m going to pull this thing up with a rope and I can stop anywhere.  If I want to model the effect of height I only have to change one variable. Hence everything is described in terms of a few variables which are meaningful when it comes time to build — height, overall length, and the length of the top driven element, angles, etc.  Besides the antenna I also modeled the feed transmission line to the transmitter.  NEC does this using an “ideal” transmission line with a specified impedance and length, the “inside” of the coax where only what happens at the ends of the line matter.  The outside of the coax is part of the antenna system itself, and includes the RF choke that defines the lower driven element of the antenna.  I then run the rest of the feed line down to ground where NEC assumes it is grounded and which could be where the radio is located.

With vertical antennas, the ground model will have an enormous effect on the level of losses, far field radiation pattern, and gain.  For comparison purposes, I use the 4NEC2’s “Average”  “Real Ground” model.  You will discover that it is difficult to get more than about 50% radiation efficiency with this model  – but that is life with a vertical antenna!

Now for some question we might hope 4NEC2 can answer.

  • How much gain will this antenna give me?
  • How high up should I pull this antenna?

The radiation pattern for this antenna at various heights is compared with the conventional 1/4 wave vertical antenna with buried radials in the figure below.


Vertical antenna radiation patterns. Conventional buried radials – blue 1/4 λ radials, red 1/2 λ radials. Vertical dipole tip at 55 feet, green; 75 feet, pink; and 95 feet, brown.

Longer radials help a little for the conventional ground-based antenna. Rasing the antenna higher improves the elevated vertical dipole.  The gain of elevated antennas is especially better for low radiation angles less than 15 degrees.  If you have a tall tree, there seems to be little reason to spend too much time burying radials in the lawn!

  • What about that choke?

The choke both defines the length of the lower section of radiator, which is the coax cable shield, and also keeps the RF from traveling down the feed line back to the radio.   The more inductance the better for these tasks, but we have to build the inductor, either air-core or with ferrite, to hang on the feed line and there will be capacitance.  I spent an afternoon with a few ferrite cores, some coax cable, signal generator and oscilloscope and can report that even good ferrite is not very suitable for this application.  Any choke I made with ferrite was lossy and had a self resonant frequency below my desired operating frequency of 14 MHz.  Much better is just a simple solenoid made from the coax feed line itself.

Unlike a simple choke on a feed to keep the RF out of the shack, the choke for this application has a much more difficult job to do.  A choke near the feed point for a normal dipole is trying to choke off the RF compared to the drive impedance at that point – which may be only a 50 to 150 ohms.  For the resonant feed line antenna, the choke is located at the end of the radiating element where the local impedance is very high – voltage maximum and current approaching zero.  This is probably the main downfall to this antenna design because it depends so strongly on getting this choke right.   If we knew the reactance of the choke we could model it with NEC.  I chose to model the inductor as a parallel LCR circuit where the tuning is not perfect for the band, but the inherent Q is high so parallel R can be >20 kΩ.  The total reactance of a tuned trap can be many kΩ if carefully constructed and tuned.

Figuring out the self capacitance of the choke coil took a little research.  Attempts at direct measurement of capacitance are confounded by parasitic capacitance of scope probes and by resonances with the physical length of coil wires and connections themselves, so that the measured resonances do not reflect the desired stray capacitance of the coil but rather the length of wires.  Internet research yielded the following formula, which seems to be plenty good for what we want.  Note that the capacitance is almost entirely just a function of the coil diameter.

Cs = 0.46 * D

where Cs is in pF and D is in centimeters, good for coils with aspect ratio 0.5< l/d < 2.0.  There were a surprising number of just plain incorrect quite recent theory papers on this subject, which lead me astray before finding this simple empirical formula discovered by Medhurst in 1947.


Feedline current with mistuned choke.

When all was said and done, I went for a about 19 feet of coax wound on a 5″ form, 14 turns.  Calculated inductance is 22.5 μH with about 6.2 pF of capacitance and theoretical self-resonance at about 13.5 MHz.
Because the choke is so critical, I decided to model several different versions and mis-tunings to see the effects.  If too far from self-resonance, the choke will leak RF current to the feed line and change the effective length of the radiator, which in turn, can degrade the SWR.

To drive this antenna several questions come up.

  • Should I use 50 ohm or 75 ohm coax?
  • What will be the SWR? How can I tune that?
  • If I use 75 ohm coax, how long should the transmission line be?

The driving impedance of a dipole in free space happens to be about 73Ω.  My transmitter would like to see 50Ω.  If I just use 50Ω cable to the antenna, the NEC model finds the expected SWR of around 1.5, which could be acceptable.  (With a perfect choke NEC could optimize to SWR 1.4 – with a more realistic choke I could only get to SWR ~1.7).  I have a large free spool of 75Ω CATV cable so I would like to show how you can win twice using  the higher impedance cable.  The basic principle is that the feed line should be an odd 1/4 wavelength multiple of the operating frequency.  Reflections from the ends of the cable will tend to cancel each other out with this special length cable.  When the cable’s impedance is the geometric mean between the input and output terminations, the cancellation of the reflected waves will be perfect!  In other words, the ratio 75Ω cable / 50Ω transmitter should be the same as ZANTENNA  / 75Ω cable, or about 112 Ω impedance at the feed point.  We can find such a higher impedance feed point if we don’t feed in the center of the dipole, but move off to one end or the other.  Notice in the model figures that the feed point is substantially below the mid-point.  The numbers I came up with were 24 ft. of #14 THHN wire for the upper section and about 9 ft. of RG-62 coax between the feed point and the choke.


Optimizer window shows rapid convergence on correct length L1 to minimize SWR.

4NEC2 can help get the correct dimensions with the “Optimizer” that is built in.  Any variables that you define, and to which you assign a value, can be used as a variable to optimize.  In this case we want the optimizer to tune the wire lengths to give us low SWR.  The model is set up with the 3/4 wave long 75Ω  transmission line.  I start off with a guess for L and L1 and let the optimizer hunt for the best values.  For this antenna, I also optimize for the antenna gain at 12 degrees elevation.  There are two solutions for SWR ~ 1 that correspond to the two equally off-center points of the dipole.  The optimizer can converge to either solution depending upon starting conditions.   The solution with the long upper wire gave slightly better low-elevation gain.  The optimizer is a nice tool, but not magic! Keep in mind the physics and know what answer you are expecting before you run the optimizer.  Choose only one variable at a time to begin with.  Remember, too , that you do not want a design that is sensitive to small parameter variations.  Experiment, but make sure you can understand the optimizer’s results based upon physical principles and you will not be led astray by numerical magic.

I just pulled this up in the tree today.  Initial tests look pretty good.  The SWR without additional tuning is around 1.2; signal levels average a few db less than the OCFD, which is what I expected.  The uniform pattern can help, however, and I already found some DX into Kazakhstan, which is in the northerly null of the OCFD.



%d bloggers like this: