Category Archives: electronics

A ground plane antenna for the VHF aeronautical band

Last month I built my first antenna. It was a ground plane antenna constructed out of a UHF female jack panel mount connector (sometimes referred to as a SO-239, like this one) and approximately 10-gauge clothes hanger wire.

A ground plane antenna made from a panel mount UHF connector and 5 segments of clothes hanger wire.

A ground plane antenna made from a panel mount UHF connector and 5 segments of clothes hanger wire.

I also used a male UHF to female BNC adapter and built my own 10 ft coaxial cable with BNC connectors. I used RG-58 and these BNC connectors because they don’t require an expensive crimp tool.

The reason I built this antenna is that for years the Saskatoon Soaring Club has struggled establishing reliable radio communications between their winch and their gliders. It’s important to mention that at Cudworth (CJD2), we use 123.2 MHz for communications. What was happening is that while the glider is on the ground, the winch can “hear” the glider, but the glider can’t “hear” the winch. A hand held radio is used in the winch and a more powerful radio permanently mounted radio is used in the glider. Testing revealed that it was likely that the radiation pattern of the antenna on the winch poorly suited for the needs of the club (more details below). The solution was to build this ground plane antenna, since they have the radiation pattern required by the club.

The old quarter wave antenna on the winch relied on using the metal roof of the cab as the “ground plane”. So long as the ground plane is sufficiently large (at least 1/4 wavelength in radius around the feed point of the antenna) the radiation pattern will be toroidal (donut shaped) and able to cover low elevation angles.Screenshot of spectrum analyzer At 123.2 MHz, a quarter-wavelength is approximately 60.9 cm (24 inches). Unfortunately, the amount of metal on the roof of the cab was insufficient (only about 20 cm). Theoretically this means that the resulting radiation pattern will be conically-tapered in elevation resulting in very little radiation being emitted at low elevation angles (low to the ground).  This document describes the problem in greater detail and shows that for real-world situations the radiation pattern is in fact more complicated than the naive description just given. In contrast, the ground plane antenna is built to include it’s own ground plane (see the 4 radials in the picture above). While searching for antenna solutions, I also stumbled upon the inverted L antenna, very cool.

Cutting the vertical wireAfter building the antenna, I needed to tune it. Thankfully, I was allowed to use a network spectrum analyzer at work to do this. The antenna started at a resonant frequency of 98.995 MHz (with the vertical wire at a length of about 30 inches). Of course, tuning the antenna to a higher frequency required me to shorten the vertical wire and the radials. It was interesting to shorten the antenna wire by 2 inches at a time and observe the resulting change to the resonant frequency measured by the spectrum analyzer. By carefully trimming the wires of the antenna, I was able to tune it to a resonate frequency of 123.7 MHz with a bandwidth of approximately 20 MHz or so.

Finally, the antenna was tested in by mounting it to the winch. This was accomplished using U-bolts, tape, and 5 feet of PVC pipe.

The ground plane antenna temporarily mounted on the SSC's winch for real-world testing.

The ground plane antenna temporarily mounted on the SSC’s winch for real-world testing.

Testing was performed by first attempting to communicate with a grounded glider using the old antenna on the winch and then mounting the new antenna and attempting communication. As before, using the old antenna, 2-way communication could not be established, but when using the new antenna the problem was resolved!

The antenna works well and club members are very happy to be able to now communicate more easily with the winch. This has made the winch launching operation safer too.


Why doesn’t my NES Zapper work?

When I was around 4-5 years old I remember playing on a NES while over at friends’ of my parents. Growing up, the classic NES titles like Super Mario Bros., Duck Hunt, Super Mario Bros. 3, etc., and now that I’m “an adult” I have my very own NES. The only problem is that I recently discovered that I couldn’t play Duck Hunt! The game cartridge works perfectly fine and I even have an old analogue CRT TV that is compatible with the NES Zapper, but for some reason (as you will learn as I did) the gun didn’t work when I would fire up a Duck Hunt session.


The first thing I did was check Google to see if anyone else encountered this problem. Most of the results pertained to issues with getting the Zapper to work with LCD TVs, (not my problem, but to understand why it doesn’t work with LCDs, you can checkout out the Wikipedia article) so I disassembled the Zapper’s connector and connected a multimeter to monitor the voltage levels when the Zapper was connected to the NES.

 nes nes_zapper

The measurements revealed that when the trigger was pulled, the voltage on pin D4 (bottom left pin of the picture where I’m holding the connector) of the NES controller connector didn’t rise to 5V from 0V; instead, the voltage would stay at 5V and very slowly decay. I decided to take apart the Zapper to see if it was the problem (and because I wanted to see how it worked!).

 nes_connector4 nes_connector
nes_connector3 nes_connector2

If you have ever used a Zapper, you may have noticed (or maybe not) how well balanced the controller feels in your hand. As you can see in the picture, the Zapper has a metal weight in the handle that helps balance the controller. There is also a lens and metal collimator in the barrel of the Zapper. Well actually the whole barrel is really a collimator and the metal collimator probably acts as another weight to help balance the controller.  Finally there is the trigger mechanism and the Zapper circuit. The Zapper circuit handles detection of trigger pulling and whether or not the user has “hit” the target in Duck Hunt.


I mapped out the circuit (very roughly…) and found that the circuit is really two circuits in one: 1) a trigger circuit with a 10uF de-bouncing capacitor, and 2) some kind of circuit that decodes input from a PD 43Pi photodiode, used to detect if a duck has been hit or not, using an IR3T07A chip (couldn’t find the datasheet!), and the output from this chip is inverted using a 2SC458 NPN transistor. This output is sent to the NES through the D3 pin.

inside_zapper zapper_circuit
zapper_circuit3 zapper_circuit2

Testing of the Zapper circuit determined that it was working. I checked the capacitors and resistors and they all seemed to be ok. I also checked to see if varying the intensity (brightness) of light on the photodiode varied the voltage output on pin D3 (the pin right above D4) and it did change as expected; the brighter the light, the larger the voltage. These results suggested that there must be a problem with the NES, so I put the Zapper back together and connected it to the player 1 port on the NES. Pulling the trigger showed that the voltage on D4 would jump from 0V to 5V to 0V as it should! So the problem was with the player 2 port on the NES. At this point, I also found a website with a nice circuit diagram and description of the Zapper gun.

The Fix


When I opened up the NES, the problem was immediately obvious. As you can see in the picture, for some reason, one of the wires wasn’t connected to the player 2 port on the NES! As you might expect, after connecting this wire to the port and re-assembling the NES the Zapper gun worked!

Now I can relive my childhood nostalgia and contribute to virtual duck murder unabated!