Here’s a picture of the transmitter I’m currently running on 630 meters.
VE7CNF 630m Transmitter
Yikes! A mess of wires! The transmitter is breadboarded while I try various ideas. I’m afraid to disassemble and rebuild it, as I’m sure the 630m band will be granted to US hams as soon as I take it apart!
The 12V 200W version of the push-pull power amp and output filter are along the top. Output power has since been increased. The bottom left breadboard has the T/R sequencer and phasing SSB modulator. The bottom right breadboard holds the DDS PCB and a microcontroller.
The transmitter is controlled from the front panel of my Icom IC-7410. It connects to the IC-7410’s serial CI/V interface and reads the tx frequency and operating mode from the transceiver a few times per second. I also connect a laptop PC for operating QRSS, WSPR, JT65, JT9 etc.
Icom IC-7410 controls the transmitter
My home-brew DDS board uses an Analog Devices AD9833, which is simple and can generate a clean x4 clock for the LF/MF band modulator from a 20 MHz crystal. The DDS chip and reference oscillator are small surface-mount parts, so I decided to design a PCB. An Atmel Atmega128 microcontroller is used to set up the DDS chip. The reference oscillator is a 20 MHz precision VCTCXO that can eventually be GPS-controlled.
Home-brew DDS board with AD9833 and precision TCXO
The phasing-type SSB modulator is based on a design by WD2XES (see http://www.w1tag.com/Phasing.htm). I designed a PCB for the audio phase shifter portion using parts of circuit designs by W3FJJ, W1VD, and KK7B (see http://webpages.charter.net/wa1sov/technical/allpass/allpass.html and http://www.w1vd.com/Phasingexciter.pdf). With a phasing SSB modulator and nonlinear power amp, I can use audio from a PC to transmit any digital mode that uses single-tone continuous-phase FSK (WSPR, JT modes, RTTY). I’d have to use a linear power amp to transmit PSK. The carrier and unwanted sideband are suppressed by 55 dB.
Home-brew W3FJJ Audio phase shifter board
I have an inefficient antenna located near trees, so I need some power to reach full EIRP. My switching push-pull “power amplifier” uses two IRFP4227 MOSFETs and is capable of 500W with a 48V power supply. It’s really an RF power supply rather than an amplifier. I use a ferrite E-core output transformer followed by a series-resonant tank and 5-pole low pass filter.
Push-Pull Power Amplifier
A large air-variable capacitor fine-tunes the series tank to peak the supply current and output power. The output filter requires many parallel film and mica capacitors to handle the RF current. The series tank capacitors and filter inductor cores get a little warm while transmitting.
The large heat sinks on the MOSFET’s aren’t necessary, as their temperature only rises a few degrees during transmit. I may experiment with making the amp run in a linear mode, then the heat sinks will be put to use.
Four cheap/free PC power supplies form the 48V power supply. Three were modified to isolate their output side from ground so they can be connected in series. Each supply has a large reverse-protection diode across its output, so supplies can be turned off to reduce voltage.
48 VDC Power Supply
I found that cheap PC power supplies create RF noise on the AC line, so I had to add AC line filtering at the main power input. I’d look for better quality supplies if I did this again. Also, it’s a lot of work to wire all this together so it may be worth the money to buy a big 48V power supply.
The antenna tuner uses a sliding variometer to resonate the antenna, and an autotransformer wound on a toroid core to match the antenna resistance to 50 ohms. The variometer uses a 3.5” diameter coil sliding inside a 4.5” coil. Coils are wound with #14 stranded insulated 600V wire.
Variometer coils before assembly
The outer pipe has a slot to allow a control handle to pass through to move the sliding inner pipe.
Assembled sliding variometer
The tuner fits into a plastic toolbox. I use an RC model servo to move the variometer by remote control from the shack.
Variometer in box with remote control servo
I have not noticed any heating of the black ABS pipe or the toolbox, but the coil wire and powdered-iron toroid core get warm after long transmissions.
There is 150ft of RG213 coax between my transmitter and the tuner. I’ve found I can tune accurately by adjusting the variometer so that the power amplifier supply current is the same value as when a dummy load is connected. To double check, I also use a “scope match” (by MØBMU, search for “VE7SL scope match”) to monitor the transmitter output voltage and current waveforms to check that the transmitter sees a 50-ohm resistive load. I can watch transmitter RF output current vary as the antenna wire swings in the trees.