Zeppelin Antenna as a 2-Band HF Dipole

Zeppelin Antenna as a 2-Band HF Dipole
by Dick Reid, KK4OBI

In the Beginning

Around 1895 Count Ferdinand von Zeppelin and Guglielmo Marconi had their amazing ideas. By 1912 Zepelin airships were beginning to carry passengers and were equipped with “Hertzian” wireless transmitters, similar to the Marconi systems on ships at sea... like the Titanic. These spark-gap transmitters produced a very broad, interference producing signal. Operating frequencies were necessarily expressed broadly in meters with 575-to-2700 meters being the “Long Wave” operating range for early Zeppelins through WWI.

There were no tuned circuits then. The spark was produced in the antenna which was was a long, weighted wire reeled out from the rear of the command gondola.

After World War I

With the coming of oscillator circuitry, frequency control became possible. Identifying a radio signal by its frequency (kilocycles, kc, megacycles, mc) rather than generally in meters length proved much more practical and useful. The rapid developments in the “useless short wave” radio frequencies (under 1500 meters) lead to vacuum tubes, continuous wave (CW) transmitters, amplitude modulation (AM), amateur enthusiasts, the concept of radio “bands” and international regulation. The addition of a high frequency (HF) radio to Zeppelin airships produced the need to match antenna and transmitter over an extremely wide frequency range. From this need came the two-wire Zeppelin antenna.

1. By the time of the Graff Zeppelin in 1936, and the Hindenberg in 1937, the 2-wire Zeppelin antenna was reeled electrically from a state-of-the-art radio room in the forward part of the luxurious 20 passenger gondola.

2. For tuning, one wire was extended 0.75 wavelengths (or multiples). Maximum length was 120 meters for use down to the 1,000-2,000 kc range.

3. For matching, the second wire was extended 0.25 wavelength (or multiples) to provided low impedance points anywhere in the high, medium and low frequency bands.

4. The main transmitter's operating range was 17.7 kc to 4.28 kc. CW and Voice.

5. The main receiver ranged from 15 kc to 20,000 kc in ten bands.

Notes:
Spark-gap transmissions were banned in 1936 in favor of CW transmissions.
In 1960 the unit of frequency CPS (cycles) was replaced by Hz (Hertz), kc by kHz, mc by mHz.

Basics: The Zeppelin Antenna

Electrically the “Zepp” antenna has two parts, a half-wave radiator end-fed by a quarter-wave matching section connected to the radio. The antenna may be in-line or in a right angle L-shape or hanging free. The end-to-end length of the total wire operates as two half-wavelengths which is, by definition, a second harmonic.

Related antennas are: J-pole, EDZ (Extended Double Zepp), Half Square, Bobtail Curtain, End-Fed Half-Wave

Figure 1
Here are the two
half-wave currents
... end-to-end.

For modeling the CF center feed point (circle) is at the low impedance/high current point at the center of the half-wave current on the matching section. (It is interesting to note that the antenna could also be fed at the low impedance/high current point in the center of the half wave radiator).

Through antenna modeling I found that when a Zepp antenna is tuned to 28.4 MHz, that frequency is a harmonic of 15.4 MHz. Presumably due to things like elevation, wire separation, velocity factor and end-effect the harmonic relationship is less than a numerical 2:1. In general this frequency ratio varies between 1.8 to 1 and 1.9 to 1 with an average around 1.85 to 1.

In this case it calculates: 28.4/ 15.4 MHz ≈ 1.84 to 1.

Graphics

Here is the sweep to 28.4 MHz @ 1.1 SWR		and enlargement of 15.4 MHz @ 1.6 SWR

Graph 1		Graph 2

Here are the two currents of the two half-waves of 28.4 MHz. The 2nd harmonic.

Figure 2

Compared to the single current for the single half-wave of 15.4 MHz. The 1st harmonic

Figure 3

At the 15.4 MHz frequency I consider the shape of the antenna as a dipole with a folded end.

Note: the antenna could also work very well as a single band, center-fed, 15.4 MHz, half-wave, folded end dipole when fed at the low impedance/high current point where wires 1 and 2 meet.

Modeling

From my Zeppelin antenna 4NEC2 model here, the following is the “Symbol Conversion file” definitions and calculated #14 wire dimensions for resonance at 28.4 MHz:

freq = 28.4
hgh = 33                                      << Feed point elevation over ground in feet
sep = 1                                        << Wire 3: Separation between parallel wires in Feet
X = 17.03781                               << Wire 2: Half-wave radiator in Feet
A = 8.62531                                 << Calculated Quarter-wave in Feet
Acf=A-sep/2 = 8.12531              << Zepp Wires 1 and 4 minus Half of the separation in Feet
Embeded Calcs = 2
Long=X+Acf = 25.16312               << Length of the long side
TotWire=Long+Acf = 33.28843     << Length of total wire; about � wave on the 20 meter band

The 14.9 MHz harmonic frequency is too high for use in the 14.00 MHz to 14.35 MHz 20 meter amateur radio band. And of course, being harmonics so you can only tune to one of the frequencies. Therefore there are no two-band possibilities.

However through modeling I find that by adding a stub (wire 5) to the end of wire 1, the stub length compensates for the shortness caused by end effect, etc.

Figure 4

This simple trick allows 2:1 fine-tuning of the 1^st harmonic to any 20 meter frequency!

This understanding belies common myths about extensions under the “J”... being. at zero potential with respect to earth… not being part of the antenna… length does not matter… no effect on radiation pattern… no RF currents if connected to the support, etc.

Stub Application for 10 and 20 Meters

As can be seen in Table 1 below, stub tuning has little effect on the SWR of the 2nd harmonic at 28.4 MHz. However, the SWR at 14.2 MHz rises because of a off-center feed situation created by the extra length added by the stub.

To get a good match for 50 Ohm coax, move the feed point onto wire #1 to fine tune the 20 meter SWR.

Table 1 Zepp Antenna for 10 and 20 meters
Changes produced by Stun Tuning and moving Feed point
Frequency	SWR	Wire 2 (X)	Wire 1 (Acf)	Stub Length	Get MHz	@ SWR =
28.4 MHz Zepp tune 14.2 MHz Stub tune	1.33 1.37	17.0378 ft. ditto	8.6253 ft. ditto	0 ft. 3.6 ft.	15.0 14.2	1.5 2.5
After feed moved	1.37	Feed point on Wire #1 at 30% to 40% from Wire #5		3.6 ft.	14.2	1.04

Now we have a completely tuneable, low SWR, 10 and 20 meter dipole antenna that is around 15% shorter than a standard 20 meter dipole.

To further shorten the antenna the stub can be bent in half to form a second stub.
Figure 5
(Wire 5)= 4.74 /2 = 2.37 ft. (Wire 6) = sep = 1 ft. (Wire 7) = 1.37 ft.
Wire 2 =16.67 ft. Wire 1 & Wire 4 = 8.195 ft. Stub=4.74 ft. Feed at 10% of Wire 1

Figure 5

For the shortest antenna, bend the stub down by the separation distance to form a second parallel matching section. Figure 6
(Wire 5) = sep = 1 ft. (Wire 6) = stub-sep = 6.84 = 5.84 ft.
Wire 2 =16.82 ft. Wire 1 & Wire 4 = 8.102 ft. Stub=6.85 ft. Feed at 10% of Wire 1.

Figure 6

Model Results

With #14 wire at 33 feet elevation, the 20 meter dipole radiation Take Off Angle (TOA) is 30 degrees.
Gain is good: 6.1 dBi compared with 7.32 dBi at 25 degrees for a standard dipole.

As seen below, the 3D radiation patterns and data, the antenna at 33 feet elevation perform very nearly the same as a conventional 20 meter dipole.

	28.4 MHz 2nd Harmonic	14.2 MHz Stub tuned 1st Harmonic
Graph 2
3D Flyover
	Two lobes at 2 half-waves elevation, 33 ft Gain: 7.77 dBi and 15 and 50 degree TOA	One lobe at 1 half-wave elevation, 33 ft. Gain 6.1 dBi and 30 degree TOA

Notes
1. The Zepp type of antenna is unbalanced and notorious for generating common mode current. It is required to have at the feed point a wide-band choke (1:1 current balun) suitable for the two frequencies. This is to prevent the coaxial cable from becoming part of the antenna which makes tuning difficult if not impossible.

2. The Zepp+Stub antenna works only where the added stub length can lower the 1^st harmonic frequency to a 2:1 ratio by compensating for the shortness caused by end-effect. For example a 6 meter Zepp dipole can easily be stub tuned longer for 12 meters but not shorter to 10 meters.
Similarly, is not possible to stub tune a 10 meter Zepp dipole antenna shorter to 12, 15 or 17 meters.

3. At this writing I have not built a HF amateur radio antenna of this type, only a vertical lab prototype in the VHF range for proof of concept.

4. When connecting a J type of antenna to coaxial cable the shield goes to the short element.

5. An arbitrary spacing of 1 foot has been used to make graphics clear for this articles. To this point there has been no discussion of the spacing between the parallel wires.

Wire Spacing

In the early days open wire feed or “ladder line” spacing was 2 to 5 inches between wires with spacing “rungs” being separated by 5 to 12 inches. We do not know if or what spacing was used by the later Zeppelins. However wider spacing may have been used because of the Low (~1 MHz) frequency carryover from spark gap transmitter use.

In the 1920’s the amateur radio adaptions of the Zepp antenna used 9 inch feeder spacing for 600 Ohms impedance. In 1928 international amateur radio bands and call signs were established. However, when the Zepp antenna was used for these harmonic bands, the high impedance at thousands of Ohms could be difficult to match. In the 1930’s the Zepp based Collins Radio “Multi-band Antenna System” used 6 inch wire spacing for 300 Ohms impedance. This is more in line with the mean harmonic impedance resistance of the harmonics which was generally under 1200 Ohms.

Coaxial cable and TV flat twin lead cable arrived in 1950’s. Since then high frequency, HF, mono-band Zepp antennas also use 450 Ohm twin lead (1 inch spacing) or 300 Ohm (0.3 inch spacing) connected to coaxial cable through a 9:1 or 4:1 balanced to unbalanced (balun) transformer.

The arrival of antenna modeling in the 1970’s opened understanding of how the Zepp works and how to adjust dimensions for 50 Ohm impedance… and for direct connection to coaxial cable. Development of the J-Pole version for Very High Frequency, VHF, amateur radio led to computer optimization of wire spacing using the formula: 22/MHz = Spacing in Feet. See Table 2 below.

Band	MHz	Feet	Inches
2 Meter	144	0.153	1.83
6 Meter	52	0.423	5.08
10 Meter	28.4	0.775	9.3
20 Meter	14.2	1.549	18.6

Table 2 J-Pole spacing based on 22/MHz= Spacing in Feet

As you can see this optimum spacing quickly outstrips conventional spacing in the HF bands and becomes impractically wide. Fortunately wire spacing is not critical for 50 Ohm designs.

Tuning

The usual tuning difference between Zepp and J-Pole antennas is the location of the feed point.

For modeling, the Zepp antenna is center-fed at the mid-point of the closed end (figure 7, wire 3). From this point to the end of the parallel section (wires 1 and 4) adds up to the quarter wave matching section. The low mpedance/high current point is in center of the end (wire 3) as seen below. Tuning is by length of wires. Notice that the RF current (color) is symmetrical and highest (red) near the feed point (circle).

Zeppelin
Figure7 Center Fed Zeppelin antenna tuning

A J-Pole antenna is also tuned is by length of wires but with both sides of the matching section are a quarter-wave in length. (Figure 8) Whatever length of Wire 4 used for connecting the parallel wires makes the total RF resonating length too long (low frequency). Tuning (raising the frequency) is accomplished by finding and connecting coax to the low impedance/high current point between the parallel wires... hence matching Zepp tuning. The location of the tuning point (wire 7) is equal to one connection wire length or more away from the closed end (Wire 4) of the matching section

J-Pole
Figure 8 J-Pole antenna tuning by shorting

The current (color) is also high (red) near the feed point (circle on wire 7), but I notice that the J-pole RF current (red) is not as symmetrical as the Zepp antenna seen in Figure 1. There is very little current (light blue) or impedance in the closed end (wires 1,4 and 5). The extra length of the closed end of a J-pole makes it difficult if not impossible to do stub tuning for dual band operation.

In fact, there is no electrical reason for Wire 4 to connect the horizontal elements. Only the Wire 7 connection is needed in a J-pole antenna. In that case stub tuning will work and SWR will be best if the feed point is on Wire 2 at 10% from Wire 7.

Stub tuning of a 20 Meter CF Zepp antenna to add 40 Meters

From my CF Zeppelin Stub antenna 4NEC2 model here, this is the “Symbol Conversion file” definitions and calculated #14 wire dimensions for 14.2 MHz resonance optimized to 7.2 MHz by Stub tuning.

freq = 7.2
hgh = 65                                                       << Feed point elevation over ground in Feet
sep = 1                                                         << Wire 3: Separation between parallel wires in Feet
X = 33.92424                                              << Wire 2: Half-wave radiator in Feet
A = 17.24027                                            << Calculated Quarter-wave in Feet
Acf=A-sep/2 = 6.74027                           << Zepp Wires 1 and 4 minus Half of the separation in Feet
Stub = 6.927005                                          << Wire 5: Tuning stub in Feet
Embeded Calcs = 2
Longwire=X+Acf+stub = 57.591515          << Length of the long side
TotWire=Longwire+sep+Acf = 75.331785    << Length of total wire

As summarized in Table 3, the 2-band results of this Zepp antenna model are good:
40 Meters at 7.3 MHz gives 1.04 SWR and 20 Meters at 14.2 MHz gives 1.07 SWR.
The radiation pattern at 33 feet elevation for 14.2 MHz is in Figure 3.
On 40 Meters that elevation is only a quarter wave high so the pattern will be oval shaped with high radiation TOA.

Out of curiosity the table compares also the theoretical and Zepp fundimental harmonics to see there was any practical use there. The answer is generally no, but possibly on 6 Meters around 51 MHz, however the SWR is high

Harmonics	Fundimental	2nd Har.	4th Har.	8th Har.
theory by antenna model	7.45 MHz	14.9	29.8	59.6
of 20 Meter Zepp antenna	7.45 MHz	14.2	28.0	56.6
SWR	1.3.	1.04	2.27	8.1
Results below are by adding a stub to tune down to 7.2 MHz and by moving the feed point to ~40% on Wire 1 for low SWR.
of 40 Meters tuned by 6.9' stub	7.2 MHz	14.2	25.4	51.0
SWR	1.04	1.07	3.9	7.8

Table 3 Effects of Stub Tuning and Feed Point adjustment, 40-20 Meter Zepp antenna

Below is the current distribution when using a stub for frequency tuning and making feed point adjustment for low SWR.

Zepp+Stub One half wave 1st Harmonic
	7.2 MHz, 1.04 SWR
Zepp+Stub Two half waves 2nd Harmonic
	14.2 MHz 1.07 SWR
Figure 9 Currents on a Stub Tuned 40-20 Meter Zepp Antenna

Below is a graphic aid to estimate the 40 meter stub length at any 20 meter frequency.

Graph 3

This graph illustrates that a 20 meter Zepp antenna can be tuned anywhere in the 40 meter band by using the tuning stub method. (Wire 5)

The two-band possibilities of a Zeppelin type antenna are:

6 and 12 meters
10 and 20 meters
15 and 30 meters
20 and 40 meters
30 and 60 meters
40 and 80 meters

In summary:

A Zeppelin (Zepp) type antenna resonates as two half waves which is, by definition, a 2^nd harmonic.
The 1^st harmonic appears at 1.8 to 1.9 ratio lower in frequency.
This lower frequency can be stub-tuned as desired to a 2.0 ratio to create a two-band antenna on any frequency on either band.
On the lower frequency, the antenna is around 15% shorter than a dipole with gain within a dB of a dipole.
A J-Pole type antenna does not ordinarily provide the two-band possibility.

Dick Reid, KK4OBI at QSL.net