[2.5 m, aluminium radiator]       [5 m, aluminium radiator]      [6m, telescopic fishing pole]   


[3m fishing pole, "Cobra" wire]      [comparison tests]     [center-loaded short vertical]


© 2009-2010 F. Dörenberg N4SPP



The 80 m KGD dipole is a bit unwieldy for my terrace and I wasn't making much progress with its pruning & tuning. At a height less than 15 m (50 ft), it should be installed vertically anyway, and that posed some problems as well.  So I decided to re-use half of the KGD and turn it into a simple short base-loaded vertical with a single radial wire: 

The radiator is also from my KGD-80 dipole experiments: 2½ m (8'2½") aluminium tubing with 7½ mm OD and 5 mm ID. I had already tapped a standard M6x1 thread in the tube ends.

The coil-core of my KGD antennas is regular 32 mm OD PVC tubing from the local Do-It-Yourself store. My antenna "mast" is 40 mm OD PVC tubing in an umbrella stand. So I glued a 40-to-32 mm PVC adapter to the coil core, to be able to simply install it on top of the "mast". I added a female BNC chassis connector for the coax and a banana plug jack for the radial. Note that the PVC adapter should not be glued until the pruning & tuning is done, so as to be able to access the BNC connector.

32 mm OD PVC sleeve and cap with various flat washers and an M6x25 bolt

As a counterpoise, I use a simple 7 m radial (household hook-up wire, "zip cord"), strung horizontally from the bottom end of the coil core. No particular reason for using 7m, it is just the length of wire that I had on hand. The top of the mast is at a height of about 2 m above ground level.

Note that a small number (e.g., 4) of elevated radials is just as effective as a large number (e.g., 120) of radials on (or in) the ground!!! Ref.: p. 32 in "An experimental look at ground systems for HF verticals", Rudy Severns (N6LF), QST, March 2010, pp. 30-33.

Antenna coil core with 32-40 mm PVC adapter, 7 m radial, choke balun, and two radiators

My original KGD-80 coils each had 112 windings of AWG #20 CuL (0.8128 mm diam). I reduced this a couple of windings at a time to 94 windings. With 0.8 mm CuL, this should only take 93 windings. According to an on-line coil calculator, the coil is now about 98 μH. I got about 14 kHz increase in resonance frequency for each reduction by one winding.

Short antennas tend to be narrow-bandwidth, and I wanted to cover both 3575-3580 and 3600-3610. As I have easy access to the antenna, and I had two 2½ m sections of Alu tubing, I decided to do the required switching by simply swapping the radiators. As shown in the antenna analyzer plots below, resonance frequency is 3568 kHz with the 250 cm radiator, and 3626 kHz with the 240 cm radiator. I.e., 58 kHz shift for 10 cm (4") reduction in radiator length.

The analyzer plots show a very nice resonance dip, with an SWR of 1.09, and Rs close to 50 ohms. Plots were made with the antenna hooked up with 13 m (40 ft) of RG58 coax and a choke balun. Adding the choke does shift the resonance frequency for some reason.

 31 December 2009: Now I'll have to put the antenna to the test, and see how it compares to my ZS6BKW multi-band dipole antenna in inverted-L configuration. Expectations regarding the efficiency of short antennas in general, and verticals with few radials should not be set too high! Anyway, I gave it a try, about an hour after local sunset. Called two PSK31-stations, both replied: PE1CDE and DO5MBB. Both at a distance of about 950 km (≈600 miles). I am pleased! To be continued! As a future modification, I'm contemplating tapping an M6 thread into the tip end of the shorter radiator tube, and use an M6x30 screw to tune the resonance frequency down when & as needed (≈ 5 kHz/cm). Alternatively, I can use a strain-relief fitting. I have described this as part of my KGD Antenna (see update of 8 Jan 2010). The aluminium tube that I use for the vertical antenna, fits tightly into such a plastic fitting. A 4 mm diameter brass rod (not threaded) can be used as capacitive tuning rod (insulate with shrink tube).

Frequency sweep plot with my miniVNA antenna analyzer - 250 cm radiator, 32 mm coil diam.

Frequency sweep plot with my miniVNA antenna analyzer - 240 cm radiator, 32 mm coil diam.

24 January 2010: the loading coil described above is actually a coil saved from my KGD-80 mini dipole experiment. For this short vertical, I ended up adding a number of windings to the coil (actually, I removed too many windings and then needed to add them back). During a Hellschreiber QSO a week ago, the solder joint came loose and the coil became uncoiled. I put it all back together again, but now the resonance frequency was some 300 kHz off. No idea why... So I decided to make a new coil, this time with a larger diameter - just for fun and out of curiosity. I used standard PVC tubing with 50 mm OD.

When increasing the coil diameter, the required number of windings is decreased by a factor that is equal to the ratio of the square of the old and the square of the new diameter (basically the ratio of the cross-sectional areas):

Based on this, when going from 32 to 50 mm diameter, the number of windings would have to be reduced from 94 to 39. This time I added a generous number of extra windings. Removing several windings at a time, I tuned the resonance frequency to 3576 kHz - just where I want it. Ended up with 42 windings: within 10% of the estimate. See table below. 


Coil windings

resonance freq (kHz)



kHz per winding






























Coil tuning data (50 mm coil diam., 250  cm radiator, 11 m coax, current choke, 7 m radial)

Obviously, when removing one or more coil windings at a time, it is hard to tune the resonance frequency to exactly where you want it. I actually wanted 3578 or 3579 kHz. So I tapped an M6 thread into the tip of the radiator tubing and use an M6 screw to adjust the effective length of the radiator (see photo below). This works quite well.

Fine-tuning is done with an M6 screw at the tip of the radiator


Per my antenna analyzer, the SWR is a little higher than with the 32 mm coil diameter. However, the SWR meter of my antenna tuner (tuner in bypass mode without bypassing the SWR-meter) shows 1:1 and my transmitter is perfectly happy without the tuner. Just a coax current choke. And this version of the antenna also works - without the coil getting warm at 100 W. Have had relatively solid QSOs with PA0KDF and PA3FFZ in Hellschreiber mode on 3579 kHz (≈1100 km, 680 miles).


The 32 mm and 50 mm coils side by side - with adapter sections to my 40 mm diam. PVC "mast"


With 32 mm diameter, my antenna analyzer claimed an Rs very close to 50 Ω, but with 50 mm diameter, this increased to 63 Ω. Not sure if this can be explained by the self-capacitance of the new coil being a couple of pF higher than that of the original coil. The new coil should have a higher Q due to the lower resistance of the new coil (using 6.6 instead of 9.4 meter CuL wire); I sort of expected to see a sharper/narrower resonance dip, but this hasn't materialized. But it works - so what me worry?!


Frequency sweep plot with my miniVNA antenna analyzer - 250 cm radiator, 50 mm coil diam.
via 11 m coax, current choke, 7 m radial)


4 June 2010: having read somewhere that adding a capacitive hat to a vertical antenna will reduce the resonance frequency, I decided to give it a try and see what happens. First, I made a capacitive hat. This is straightforward. I got two 3 mm threaded rods of 1 m in length, and cut each rod into three pieces of 33 cm (≈1 ft). I then took an M6 screw with hexagonal head (unlike the M6 tuning screw described above), drilled holes through all sides, and tapped an M3 thread into each hole. Then I screwed the rods into these holes, and fixed them in place with an M3 nut. That's all!


Results are twofold:


it works as advertised: resonance frequency went down from 3574 to 3176 kHz - a whopping 400 kHz! SWR was reduced insignificantly: from 1.42 to 1.37.


This hat weighs 95 grams (≈ 3.4 oz). This is (too) much for my rather flexible aluminium radiator. The slightest breeze will make this antenna sway a lot. On a more rigid antenna, this would be no problem.


The prepared M6 screw and the six rods

Close-up of the hub of the capacitive hat

The hat atop the vertical antenna

23 February 2010
: have made a number of QSOs with th 2.5 m version if my antenna, from southern France to southern Germany (750 km), and to the northern part of the Netherlands (1100 km). Not bad for the size of the antenna! But a longer vertical should be even better. So I am simply going to extend the radiator with a second section of 2.5 m Alu tubing. For sure, this version will not be storm-proof - at all!

I have tapped an M6x1 thread into the ends of both of the tube sections. The threads are 2.5 cm (1") deep. I cut a 5.3 cm (≈2") piece off an M6x60 bolt, put two M6 nuts at the center and tightened the  nuts against each other.

This will be the link between the two tube sections. The tube sections are simply screwed all the way up against the nuts. 

Well, in principle, the idea was not (so) bad. But this small-diameter aluminium tubing is so flexible, that it will absolutely not stay upright. The radiator now is 5 m long, but even without wind, it bends over such that it is still only 2.5 m high! Oh well, should have seen that coming, hi! Now I have to go back to the Do-It-Yourself store and see if they have 2,5 m Alu tubing with a larger diameter and a thicker wall. I'll use that for the bottom half. To be continued...

7 March 2010. The 5 m tall aluminium vertical clearly was not a success, to put it mildly... But an experience is only a bad one, if you don't learn from it. I need a long tube, light weight, and fairly stiff. I have been very pleased with the 3 m (10 ft) telescopic fiberglass fishing poles that I used for my 20 m mini-dipole, mini-vertical, and mini-yagi-beam. So I decided to get a longer one. Do not use a carbon fishing pole: besides being much more expensive, they are also conductive.

Went to a local sporting goods store (Decathlon, present in many European countries, Russia, China, and Brazil) and checked out the fishing supplies department. I found telescopic poles ranging from 2.5 to 9.5 meters, and prices from €10 - €160. I settled on a 6 m pole for a reasonable €14 (≈ $20). The model name is Telepocket S 600; it weighs 530 gram (1 lb 4 oz), and collapses to a very compact 58 cm (1 ft 10").  The base section of this pole fits nicely in a 40 mm OD PVC pipe.

My antenna "mast" is 2 m of 40 mm OD PVC pipe, stuck into an umbrella stand on my terrace. I made a fishing pole stand with integrated coil core, also out of PVC.

From left to right in the picture above:


40 mm to 50 mm adapter


50 mm sleeve (to connect two sections of 50 mm OD pipe)


50 mm OD pipe


40 mm sleeve (to connect two sections of 40 mm OD pipe). This sleeve has three notches on the inside, at half height. These prevent insertion of 40 mm pipe more than half way. As it happens, the base section of the fishing pole has a diameter that snugly fits between these notches. Perfect, so the antenna will wiggle around in its stand!

All very inexpensive at the local building supplies store. The 40 mm sleeve almost fits into the 50 mm pipe. I heated the end of the 50 mm pipe by sticking it in boiling water for a minute or two. The 40 mm sleeve could then be inserted without any problem. I let it stick out a little over 1½ cm (5/8"). I also glued the 50 mm sleeve onto the 50-to-40 mm adapter.

I did not glue the 50 mm pipe into the 50 mm end of the adapter. The fishing pole has a screw-on cap at the bottom, to prevent the 11 sections of the pole from falling out. This cap does not fit through the 50 mm pipe. So the pipe has to be slid onto the fishing pole from the tip of the latter. The cap does fit into the adapter piece without a problem, and cannot fall out the 40 mm end of the adapter. If will not glue the pipe into the adapter sleeve, until the loading coil is tuned. Maybe I will not glue it in at all - it will not be installed on my terrace permanently.

I also used a short section of 32 mm OD PVC pipe, and glued that into a 32 mm hole in the 50 mm pipe. I shaped the inserted end of this piece such that it conforms to the inner diameter of the 40 mm pipe, and does not touch the inserted fishing pole. I installed a BNC jack (coax) and banana plug jack (for the radial wire) into this 32 mm pipe.

With the PVC stand and base, the overall length of the collapsed antenna is 64 cm (2 ft 1"). Total weight is 930 gram (≈ 2 lbs).

I used a 1 mm bit to drill several sets of holes for the coil. The coil is tuned by removing windings from the top side of the coil.

The radiator is simply 5½ m (18 ft) of multi-strand AWG #22 wire (Radio Shack hookup wire, item 278-1218, 90 ft for $6.60). One end is soldered to the top end of the loading coil. The wire is wrapped 8-10 times along the length of the fishing pole, to keep it close to the pole.  The end of the wire is attached to the tip of the pole (with a small alligator clip).

I re-used the 7 meter radial wire from my other short verticals. Tuning was done with 10 m coax (40 ft) and a coax choke at the transceiver end. The antenna is installed 2 m above the terrace.

I started out with a coil of 47 windings of 0.8 mm CuL on the 50 mm pipe - a couple of windings more than what I ended up with for the 2.5 m vertical), just to be safe. I reduced this in two steps to 30 windings (estimated inductance: 45 μH). The resulting resonance frequency is a little higher than my target (3579 kHz). The frequency shift per removed winding was a little higher than expected... The bandwidth (SWR = 2) is about 95 kHz. The SWR at my target frequency is very easily reduced to 1:1 with my ATU.

Coil windings

resonance freq (kHz)



kHz per winding















The frequency sweep plot of my miniVNA antenna analyzer is shown below.

I tested the antenna tonight (19:30 local time) by checking my signals via a Web-SDR in The Netherlands (≈ 1100 km). This screenshot from my IZ8BLY Hellschreiber program shows my signals after delays through the Web-SDR. Also had a 59 QSO with PA0KDF.

Further tests remain to be done, also in direct comparison with the 2.5 m vertical. I may also experiment with a longer radiator wire that is wrapped 40-50 times over the length of the pole, rather than just 4-5 times.

12-Mar-2010: Today I tuned the antenna from 3605 down to 3570 kHz, by using a radiator wire of 5m65 instead of 5m50; about 15 kHz for 10 cm (4 ").  I also experimented with 5m60 3-conductor wire (AWG #20)that I use in my linearly loaded Cobra dipole. As it is quite a bit heavier than the AWG #22 wire, I had to tape it in place at several spots along the pole. The resonance frequency dropped considerably, confirming that the linear loading does work as intended: it makes the radiator appear about 15-20% longer than its physical length. As I still have a couple of 3 mtr (10 ft) fishing poles from my 20 meter center-loaded antenne experiments, I may try a 3 mtr base-loaded Cobra vertical next.


resonance freq (kHz)



AWG #22











Single-wire and 3-conductor wire in Cobra configuration

3-conductor wire (AWG #20 rotor control cable)

Frequency sweep plot with my miniVNA antenna analyzer -
550 cm wire
via 11 m coax, current choke, 7 m radial)

The 6m vertical installed on my terrace

Did a quick check with 20 mtrs of 450 ohm ladder line instead of coax, to see if the antenna holds any promise of being useable as a multi-band. Used a choke balun at the TRX, and tried with and without a 4:1 balun. Results are tabulated below, and the associated analyzer plots are shown below the table. With the ladder line, the resonance frequency in the 80 mtr band is about 100-150 kHz lower than with the coax.

resonance freq (kHz)




Xs   resonance freq (kHz) SWR



3472 3.0 147 14   3432 1.5 45 17
5105 3.6 153 63   5067 1.3 43 21
7380 1.4 36 14   9067 1.4 68 15
9190 5.5 116 131   12040 1.5 37 16
11968 3.6 54 71   17800 3.7 40 62
13208 1.4 64 16   22772 8.3 73 152
17550 1.35 52 16   27542 2.2 52 41
22500 2.19 31 26          
27788 1.8 45 25          

Resonance dips with ladder line

(left: without 4:1 balun, right: with 4:1 balun)

Frequency sweeps with ladder line

(top: without 4:1 balun, bottom: with 4:1 balun)

today I built and tuned yet another short vertical for the 80 m band. This time I used a 3 m (10 ft) telescopic fiberglass fishing pole. The radiator is 2m7 (9 ft) of 3-conductor antenna rotor control cable (20 AWG,  Radio Shack c
atalog nr.15-1150). The wires of this cable are connected into a zigzag configuration, like I used for my Cobra dipole.


3-conductor wire in "Cobra" configuration

Again, this antenna is base-loaded. The loading coil is wound on an 18 cm section of 32 mm OD PVC. The coil core is slid all the way down on the fishing pole, until it rests on the end stop of the pole. This is then pressed into a 32-to-40 mm PVC adapter piece. The adapter is mounted onto my 2 m "tall" antenna mast of 40 mm OD PVC tube. To prevent the pole from falling through the adapter piece, I cut the top of a 32 mm PVC end cap, and glued that into the adapter piece.


The lower end of the fishing pole and PVC pieces of the stand
(18 cm PVC with 32 mm OD, 32-to-40 adapter with 40 mm sleeve, 32 mm end cap)

PVC disk sliced off the end cap

Tuning was done with my miniVNA antenna analyzer. The antenna was hooked up in my standard configuration: 11 m coax, a current choke at the analyzer, and a 7 m horizontal radial. Tuning data is captured in the table below. SWR=2 bandwidth is about 45 kHz (more if using an ATU to reduce the residual SWR to 1:1).


Coil windings

resonance freq (kHz)


Rs /Xs

kHz per winding
97 3246 1.39 65/11 -


3438 1.42 67/10 19


3554 1.38 68/5 19
81 3564 1.41 69/6 10

Coil tuning data



tonight I ran a comparative test with the 2m5 (aluminium radiator), 3m (3-conductor "Cobra" wire radiator), and 6m vertical (single-wire radiator) antennas. Tests were completed within 20 minutes, from 21:20 - 21:40 local time. Transmissions were with 50 watt on 3579 kHz. Testing was done in Hellschreiber mode (IZ8BLY software), by transmitting the characters "1 2", and using a remote receiver to check the signals. The remote receiver is this great Web-SDR in the east of the Netherlands. It is located at 935 km (580 mi), due north of my QTH in the south of France. The signal processing delays in the SDR plus the internet delays added up to 2 sec. This allows me to listen to almost 2 sec worth of my own signals after completion of a transmission and switching back to "receive". Each screenshot below represents a period of about 2 minutes. This qualitative test did not show significant differences in performance between the antennas... The tests do show that all three antennas get a decent signal out (at my location), esp. when considering their size!


I also ran a brief comparative test between the 2m5 and 6m verticals a  week ago in PSK31 mode with DF7XH (770 km, 50º to the northeast). The 6m did show a small-but-noticeable advantage. A similar test  (3m vs. 6 m) a week or two later with PA0KDF also gave the 6m vertical an advantage.


Top to bottom: the base of the 6 m, 3 m, and 2.5 m antenna


Transmission of "1 2" via 2.5 m vertical, and reception via remote Web-SDR

Transmission of "1 2" via 3 m vertical, and reception via remote Web-SDR

Transmission of "1 2" via 6 m vertical, and reception via remote Web-SDR

On of the prime design parameters of loaded antennas, is where the loading coil is placed between the feed point and the tip of the antenna. The above short vertical antennas are all "base-loaded": the loading coil is placed at the base. i.e., at the feed point.  From a construction point of view, this is the easiest and most robust: the higher the coil is installed, the more top heavy the antenna becomes, and the more it sways in the wind. With material left over from the above designs, I have made an other short vertical for 80 mtrs: a  4 mtr center-loaded vertical. It is described here.


   top of page  button - radio
button - home