I am getting back into Ham Radio, inspired by the incredible response to the Codec 2 project from the Ham community. In my home office, I had trouble listening to the local 2m repeater due to all the high speed digital kit I have. So I decided to put up an external antenna.
Through the wonders of the Internet I read about this antenna on brainwagon, which lead me to the Flower Pot Antenna web site of John, VK2ZOI. Check out John’s site for construction details.
To save some coax I just soldered a scrap of speaker wire cut to 457mm (ish) to the end of my coax, rather than using the coax inner. I used a 40mm PVC tube as that was what I had laying around, with 4 turns of coax for the RF choke. For an initial test I taped the antenna and choke to the outside of the PVC tube with electrical tape.
The SWR indicator on my FT-817ND was dead flat (perfect) between 144 and 148MHz. Signals received from the repeater are pinning the signal strength meter, and no more digital noise. Not bad for 10 minutes work! Nice antenna design John!
I put the antenna inside the PVC tube, fitted the end cap and popped it up on my roof. The grey box below is a pre-production Mesh Potato co-located on the same mast, used for my local mesh network and to experiment with Village Telco hardware and software.
How it Works
Here is my explanation of how this antenna works. I might be wrong, but it was fun thinking about it.
A regular dipole has two pieces of wire a quarter wavelength long, fed by some coax in the middle:
Now, imagine rotating the dipole so one of the wires is next to the coax:
Now the clever part. If we remove the wire, RF currents will flow along the outside conductor of the coax. So the outside of the coax will form one side of the dipole, while the inside continues to act as a transmission line.
Now if we add an RF choke a quarter wavelength along the coax, it will stop the RF currents at that point, forming the other half of the dipole. Without the RF choke the RF currents would keep on flowing down the coax, and that half of the antenna wouldn’t be the correct length. So above the choke, the coax outer is an antenna, beneath the choke it’s just a plain transmission line:
A few turns of the coax forms the choke. Too many turns and the effects of the inter-winding capacitance will be larger than the inductance and it won’t be an effective choke. John has a table of the self resonant frequency of various turns and PVC pipe diameters near the bottom of this page.
I have also seen this type of antenna described a resonant feed-line (RFD) antenna in my 1995 ARRL handbook, but scaled for the HF (3-30MHz) bands. All the dimensions are just scaled up for the longer wavelengths, including the RF choke. Because of it’s length (e.g. 40m at 3.5MHz), the antenna is strung horizontally instead of mounting inside a PVC tube.
Wifi Router Sleeve Dipoles
This got me thinking about Wifi router antenna design. Many routers use a sleeve dipole design, here is a picture of the internal construction from this Marty Bugs article:
Very similar to our end fed flower pot design, scaled down to a quarter wavelength at 2.4GHz. But where is the RF choke? And why do we need the sleeve, why not just use the coax outer? Here is my explanation….
The RF choke in the flower pot presents a high impedance to RF currents, which stops current flowing further down the coax outer. However RF chokes are more difficult to build at 2.4GHz, e.g. a choke made from coiled coax would have a self resonant frequency too low to be useful.
So rather than a high impedance in series, how about a low impedance in parallel with the coax outer? This is the function of the sleeve. For 2.4GHz energy, it “looks” like a short circuit, diverting the RF current away from the coax outer.
The sleeve is a quarter wave section. Through the magic of transmission line transformers it forms a transformer or balun. It’s operation is decribed here on Wikipedia. I must admit this infinite to zero impedance transformation sounds like magic to me. I need to read up on transmission line transformers again to fully understand this. But for now I will just accept that it presents a low impedance for the RF currents, preventing them flowing down the coax outer.
A perfect earth has a low impedance, so I guess “electrically” the sleeve looks like a very nice ground plane. So perhaps we are effectively forming a quarter wave vertical antenna over a perfect ground plane.
I am writing all this down to explain it to myself. If any one else has alternate explanations, or I have made any errors, please feel free to correct me!
Glad that some links that seemed interesting to me were useful to you David. I’d seen the coax dipole idea before, I just really like VK2ZOI’s stellar write up. I’ve been bookmarking a few related sites, such as this cool one for 10Mhz, but I hadn’t really thought of using the idea on the microwave bands. Thanks for more to think about!
I can recomend the book “Microwave Antenna Theory And Design” edited by Samuel Silver to anyone interested in wifi antennae constuction.
Here is an alternate explanation for how the sleeve works:
The unsheilded conductor (formerly the center conductor of the coax) and the sleeve form the radiating dipole. This works like any dipole. A dipole does not need a ground plane to work – the opposite ends of a dipole are at opposite potentials, and this difference in potential creates the propagating transmit signal.
The excitation for the dipole comes from the driving coax right at the middle of the dipole which is normally where dipoles are driven. The driving coax has a potential across the inner conductor and the shield. The shield connects to the sleeve at the center of the dipole, and the inner conductor connects to the unshielded conductor also at the center of the dipole.
This should all make sense, except that the coax is inside the sleeve. The coax shield is supposed to remain at zero volts (ground), yet the sleeve, which is one half of the dipole, should have an AC voltage on it. This is where transmission line theory enters.
The coax shield acts as an *inner* conductor inside the sleeve, and forms another coaxial transmission line. This transmission line is open at one end and shorted at the other end. The shorted end is at the center of the dipole where the shield is deliberately connected to the sleeve. Further, this transmission line is, by design, 1/4 wave long.
According to transmission line theory, a short at the end of a 1/4wave transmission line looks like an open at the other end. This means that where the coax enters the sleeve, there can be a voltage difference without an associated current flow.
Thus the driving coax shield can stay at zero volts and have no shield currents induced by the voltage swing on the sleeve that is one half of the dipole.
So in summary, the sleeve is one half of the dipole, and the unsheilded conductor is the other half. The driving coax drives the dipole in the center. Because the sleeve and the coax shield form a 1/4-wave, shorted transmission line, an open appears at the coax entry point. Thus the dipole voltage of the sleeve does not induce a current in the coax shield.
I hope this makes sense.
Best regards,
Mitch
Hi David,
Very interesting article. Suppose one made an S or a loop in the bottom half of the antenna, do you have an idea how its performance would change?
My interest is in figuring out how could make a 121.5Mhz (aviation comm antenna) in a small space.
best regards,
Naresh