Pages

Tuesday, September 4, 2012

The Turnstile antenna

omnidirectional and horizontally polarized VHF antenna

Some time ago my antenna rotator got faulty and I replaced temporally a 9 element Tonna with a small Diamond X-50 vertical antenna. This antenna works nice, but is very limited in tropo VHF DX work because its vertical polarization. Of course sometimes 2m opens widely and you can work some tropo DX with a vertical antenna, but of course it's not a every day occurrence... neither a every year occurrence.

Once inspected, the rotator was dead, so I decided to try a small horizontally polarized omnidirectional antenna. There was some antennas of this kind available, so I started a small research to decide what antenna to build.

The easiest antenna is the closed dipole, or halo. But this antenna have some directivity. To reduce it, the dipole must be made small and must be loaded with a capacitor at the open end. You can get very high voltages on this capacitor, even working with low power. Also the antenna has a very narrow bandwidth and can be weather-sensible. So I discarded it.

The next antenna were the so called Cloverleaf, or Big Wheel antenna. It was very attractive, but somewhat large for me. This antenna would be the chosen one for the 70cm band.

Another option was the Egg beater antenna, or the K5OE's variant Texas Potato Masher. This antenna is used heavily on satellite work. In practice, it has a problem: It needs a section of 100 or 92 ohm coax cable. It's very easy to find 50 or 75 ohms cable, but extremely difficult to get other impedances. So I discarded also this antenna.

But the Egg beater antenna is a improvement of another antenna: The turnstile. This antenna is no more than two perpendicular dipoles dephased 90 degrees to produce an omnidirectional pattern. It also seems to be very easy to construct, so this was the antenna chosen.

The antenna:

The antenna is made of two resonant dipoles placed perpendicularly and phased with a quarter wave coax run to produce a 90 degrees phase difference between the two dipoles. I modeled the antenna in MMANA software, with great success:


The turnstile antenna in MMANA. As you can see it is almost omnidirectional and the main lobe is pointing to the zenith. The gain at the horizon is -0.89 dBi

The simulation was perfect, but the numbers were not impressive. The antenna has a main lobe pointing just upside and the horizon gain, the one used at DX work, is minus 0.89 dBi. Add a coax cable to your transceiver and you will get easily a -3dBi (or even minus!) antenna gain. Something just a bit better than a rubber duck antenna.

Anyway, I decided to build and test the antenna, because:
  • When you have real tropo conditions, signals are strong, so maybe you can loss some dB and still make some interesting contacts
  • This is a very common antenna for beacons. And beacons usually run low power. The cluster is full of long range receptions of these beacons, so maybe I could achieve the same results.
  • The most important thing for any antenna in tropospheric work, is the horizontal polarization.
In resume, A working VHF antenna, really far away from a 17 element yagi, but a very capable antenna.

The making-off:

The antenna was made with a small length of high pressure PVC tube 35mm wide. At one end, four holes were made to hold the four dipole arms. The arms were made with aluminum tube. The arm had a headless screw at one end. This screw pass through the PVC holes and a pair of nuts secure it. See the drawings for more clarity.






The phasing line was made with a small piece of RG-59 coax cable (75 ohms). Both dipoles where connected in parallel with a (electrical) quarter wave length of coax. Most RG-59 cables have a factor velocity of 0.66. But surprise!!! I have found some cables marked as RG-59, with a different dielectric, that exhibits a 0.78 velocity factor. If unsure, you can measure it easily with a load and a SWR meter.


The dipole length was calculated with MMANA to be resonant. The total length (including the gap at the center) was 97.4 cm. The RG-59 coaxial pashing line was cut at 34.3 cm. Please, note the only piece of RG-59 in the antenna is the one that connect both dipoles . All other cables must be 50 ohms cables if you don't want some headache...

The antenna on the air:

It's a really discreet antenna. The antenna was mounted on July 1, 2002. It's located about 40 - 45 meters over the street level, and it was connected to the transceiver with 35 meters of Aircom Plus. The rig was a Kenwood TR-751, rated at 25 watts, so the EIRP calculated was only about 14 watts. Again the numbers are not impressive.
If you test the antenna with the usual (vertically polarized) signals, the antenna does not work at all. Signals are low, and you will lost many distant repeaters. Local ones will be noticeably weaker. This was my first impression of the antenna. A deaf antenna. But just a small experiment with a local ham and a portable dipole, demonstrated that the antenna was working horizontally.


After some days of testing the first concussions arrived. The antenna works nicely locally if your QSO partner works also with horizontal polarization. Working with tropo signals, the antenna definitively works. I could made some interesting contacts for me, even in meteor scatter. Maybe for you these distances are not impressive. Spain has many mountains, so working at VHF/UHF bands is very very difficult here.


Station Distance Mode
EA3BB/P 421 Km SSB
EA5DGC/P 331 Km SSB
EA6IB 478 Km SSB
EA5AYG 304 Km SSB
EB5EEO 352 Km SSB
EA5GLN 311 Km SSB
EB4FVE/P 319 Km SSB
EA3DXU 520 Km JT44
S52LM 1542 Km FSK441

The antenna works nicely on tropo, but it does not suffice with FAI signals. It works really NICE with sporadic E signals, and also for satellite. In fact, my personal record of distance in sporadic E was made with this antenna, 2185 Km with SV-land. I must admit I was very happy with this antenna. Later, I acquired an FT-817, only 5 watts. The antenna also worked nicely with this low power transceiver. In fact, my record in Sporadic E was made with this rig.

Conclussion:

I used this antenna for two years. It has two main disadvantages: It has a very low gain and, very important in urban areas, the antenna is somewhat noisy. It's omnidirectionality makes it a noisy antenna. Anyway it is a very interesting antenna. If you have space problems, and yagis are a no-no, try this antenna. You will get surprised!

9 comments:

  1. Hola!

    Soy un SWL con QTH habitual en EA3, en concreto JN11ai, cerca de BCN. Aunque también disfruto escuchando la HF (utilizando dos receptores, un Sony ICF-SW11 y un Tecsun PL-600 que van a las mil maravillas), mi pasión real son los satélites (mediante un escáner IC-R5 que también va fenomenal). Como de momento no instalaré un rotor (aunque esté pensando en hacerme uno basado en Arduino y con dos servos de avión R/C de alto torque, para algún tipo de Yagi ligera... pero eso en un futuro) estoy pensando en la Turnstile como antena de base para RX los sats, y también para RX "cosas" terrestres en SSB/pol. horizontal en las bandas de 2m y/o 70 cm. Pero antes de construirme una, tengo un par de dudas:

    -¿La Turnstile resuena doblemente en 2m/70 cm? Creo que un dipolo si, pero que la Turnstile no por culpa del latiguillo de enfase. ¿Es así?

    -La Turnstile son dos dipolos enfasados, entonces ¿es recomendable poner un balun en cada dipolo? ¿o uno en la bajada "general" tras el enfase? ¿O usar baluns sólo complica las cosas sin ninguna mejora real en el rendimiento?

    En fin, esto es todo... Saludos!

    David (de momento, sin indicativo, quién sabe si en un futuro...) :P

    ReplyDelete
    Replies
    1. ¡Hola David!

      Sí, la turnstile "resuena" también en 432 MHz, ya que el latiguillo de enfase, que es 1/4 de onda en 144 MHz, en 432 es de 3/4 de onda (1/4 + 1/2) por lo que sigue cumpliendo su función.

      Pero resonar no significa funcionar, y menos óptimamente. Como el diagrama de radiación de un dipolo de 144 en 432 está fuertemente distorsionado, el enfase de dos dipolos produce otro diagrama de radiación fuertemente distorsionado y muy poco práctico, tanto para comunicaciones por satélite como para trabajo terrestre: "Funciona", no da ROE, pero no "rinde".

      El uso de balums en las turstiles de VHF/UHF no suele ser buena idea, pues pueden introducir pérdidas nada despreciables. Además la impedancia resultante está lo suficientemente cercana a 50 ohmios como para que no sea ningún problema en el trabajo habitual.

      ¡Saludos!

      Delete
    2. Bien, pues entonces voy a tener que hacer dos Turnstiles, la versión de UHF y la de VHF. Las dejaré sin balun.

      He cortado un tramo de tubo de PVC en 2 mitades, así que haré una antena por mitad. Eso, junto con alambre galvanizado de 1,35 mm de diámetro, un poco del típico coaxial de TV de 75 ohmios (ojalá el FV sea de 0,88 -que creo que es un valor típico en estos coaxiales- porque no tengo instrumental para medirlo) y unos tramos cortos de RG-58, de no más de 60 cm cada uno (que no creo que atenúe mucho la señal aunque sea RG-58) serán los materiales para hacer tanto una como otra antena.

      Una última cuestión: ¿qué cable usar para bajar la señal del satélite? Mi intención es hacer una instalación semipermanente (desmontable y trasladable) en el jardín, usando un mástil telescópico de fibra de 12 m. de Spiderbeam que ya tengo (enderezado con un buen sistema de vientos mediante algún tipo de cuerda resistente a la intemperie, porque sino se dobla como una caña de pescar). Midiendo me salen mínimo unos 24-25 m. de coaxial para bajar del mástil y llegar a casa. No sé si un RG-213 sería suficiente o si sería necesario algo mejor (tipo Aircell 7 por ejemplo), especialmente teniendo en cuenta que las Turnstile van justillas de ganancia hacia el horizonte como comentas en la entrada. En definitiva, andaría buscando algo con una buena relación dB/€ para la longitud que tendría que usar, ya que, mi exigua economía de estudiante no me permite grandes desembolsos... :)

      En fin, a ver cómo sale la cosa al final. ¡Saludos!

      David.

      Delete
    3. En el cable nunca hay dudas: El de menor pérdidas que te puedas permitir. ¡Suerte con la antena!

      Delete
    4. ¡Muchas gracias! Ya comentaré los resultados e impresiones en mi blog (http://radioscanning.wordpress.com/) :)

      ¡Saludos! David.

      Delete
  2. try stacking a pair for some gain

    ReplyDelete
  3. Olá.

    Qual a medida de cada elemento?

    Obrigado

    ReplyDelete
    Replies
    1. From the text:

      The total length (including the gap at the center) was 97.4 cm.

      Delete