Pixie 2 Transceiver

 I had a lot of fun making the video series on the Forty-9er, so decided to do something similar with another cheap eBay kit I had in my junk box. I found a Pixie, so the next transceiver was decided. This radio is simple to the point of being almost useless, but it is an interesting design.

The Pixie2

This series I broke into three videos:

• Audio Amplifier
• Oscillator
• Mixer/PA

As before, there's also an activation video where I put it on the air.

Unlike the Forty-9er, however, I wrote a document that explains things a bit better. And it has the space to go into the math. Everyone could use some more math.

 

Hope you enjoy!

The Pixie Transceiver

On a whim I bought a cheap Chinese pixie kit from eBay. These are cheap kits, less than five dollars, that generally put out less than 1 watt. All that arrived was a static bag with the pcb and parts. There was a schematic on the eBay page, and I was able to assemble without much trouble. One of the capacitors was the wrong value, but I installed it anyways, since I didn't have a spare of the marked value. I didn't install the power LED or the buzzer. I don't think that this will be a problem. I attached a 9V battery and an RCA connector for the antenna and was ready to test.

Pixie, ready to be tested.

Attaching a dummy load, I keyed the transmitter. There was a click in the headphones, and the noise was muted, so it appeared that the transmitter was working. My dummy load, from QRP Guys, has pins that can be used to find the power. My calculations showed that the pixie was putting out about 200 mW. This is much less than the 800 mW at 9V the website claims.
I connected my antenna and immediately heard a number of stations. Tuning around with my KX3 I discoverd that they were up to a kilohertz away from 7.023, the crystal frequency. The receiver is wide open.

In an Altoids tin, for size comaparison. Plenty of room left for a 9v battery.

After thinking a bit, I decided that it would be easier to experiment if the radio wasn't soldered into a PCB, so I ordered two more from ebay. They're so cheap. Once they arrive, I'll assemble them on a breadboard so I can poke around with an oscilloscope and change the circuit easily. Then, one will get mounted in an Altoids tin to be brought out into the field for some QRPp fun. Stay tuned for updates when the other pixies arrive.

WSPR and Antennas

Last fall I purchased an Ultimate 3s WSPR/QRSS transmitter kit from QRP Labs. I bought the U3S, GPS module, receiver module and case, for the 30m band. Recently I got around to finishing the transmitter half and set it up for WSPR. Watching the spots come in is a lot of fun, seeing how far I can get with only 23dBm (thats 200mW for those who aren't engineers). Looking at the map of spots, it appears that my antenna might have nulls pointed at New England and the Carolinas. Maybe that's why I've yet to work any states in those regions. I have not seen many spots from Asia or the pacific, and have seen none from Europe, South America or Africa.

24 hours of spots of KB1KXL, ending 0020 UTC 2/8/2017.

I think that my antenna is to blame for the lack of results in those directions. It is a low (~15 feet) doublet, cut for 40m, and in an inverted-V configuration. I'm feeding it with 300 ohm twin lead, which has to pass through a metal window frame. I also have to keep the feedline low in some rose bushes until it gets to the antenna tree so that it can pass under the power lines. Also, one of the legs passes through a few trees. The tuner in my KX3 is able to match it on all bands 80-6. My guess is the low height and all of the compromises I had to make in order to get it up are limiting the range and ability to hear dx. To upgrade I'm going to try running some small coax through the window to a balun outside, under the theory that it will reduce losses while going through the window. A more difficult upgrade is to get a tall pole to support the center of the antenna.
Using the online VOACAP modeling for coverage area, it looks like I can get some improvement from going from a dipole at 5 meters to one at 10 meters.

Dipole at 5 meters (17 ft)

Dipole at 10 meters (33 ft)

 These plots were generated on 2-3-2017 at 0100 UTC, location Sacramento, 5W, CW, on 40m. All others settings were left at the defaults.


At some point I'll switch the Ultimate 3S over to QRSS and see how that does. I'll post an update with some results when I do. Next I need to get the Raspberry Pi set up with a sound card so I can have my own WSPR/QRSS grabber...