Today, I received a package in the mail from a fellow ham who was getting rid of some gear. Specifically, he’d put together most of the parts to put an Arduino-controlled VFO on one of the cheapie Chinese Forty-9er kits, like Farrukh Zia did in a QST earlier this year (his kit is available online.) The package I received had the Zia board, an Arduino Nano, an AD9850 DDS module, a rotary encoder, and the Forty-9er kit itself, already assembled. I do want to someone assemble the kit as described in QST, but for the moment, what I’d really like to make is an antenna analyzer.
The scope of what constitutes an “antenna analyzer” is wide – from my rinky dink setup at the bottom end to 4-figure solutions at the high end. Really, what I’ve put together is andirectional coupler with 50-ohm terminations. The fancier analyzers will tell you not only the actual impedance of the antenna (or anything else on the output port), but also the magnitude (and in some cases the sign) of the reactive component of the impedance. Neat stuff! For my purposes, a simple Arduino-driven device that measures relative SWR will be good enough for my purposes. (Dan, KB6NU, has a helpful write-up of some of the mid-range options.)
I’ve been looking at K6BEZ’s plans for an Arduino-based analyzer for a long time – really, the key piece I thought I was missing was the AD9850 – and this seemed like a good time to give it a try.
I had an SWR Bridge Kit from KitsAndParts lying around, and I thought this would be a good time to put it to good use. The kit is simple directional coupler, with a couple diodes and capacitors in each line to hold a peak DC level. The coupler is meant to drive a pair analog meter movements.
To adapt this coupler for use, I first simply attached each output to the Arduino’s analog inputs. This went quite poorly – the charge building up on the capacitors on each side of the coupler has no where to go, and so rather than giving an instantaneous peak voltage, the analog readings just kept creeping up and up. A 68K resistor from each output to ground (facilitated by a small breadboard) solved that issue.
The main issue now is one of scale. The AD9850 Module itself puts out a little less than a milliWatt at most, deceasing with higher frequency outputs. The K2ZIA module has a little two-transistor cascade amplifier to help compensate for this, but that only pumps the output up to 14 or 15 dBm at 1 MHz, with decreasing returns at higher frequencies. Given that the directional coupler has a “single turn” (1 wire) pickup through the center of each toroid, and each toroid has 12 turns, we can surmise that less than 1% of the energy going into the coupler is being diverted to measurement functions. with the rest passing through to the antenna.
This low amount of coupling would normally be desirable, but this time, that means -5 dBm presented to the FWD measurement port, which just isn’t enough power to overcome the forward voltage of 1n5711 diodes in the coupler. (I measured mine to have a voltage drop of 0.323V. That’s which a $50 multimeter, so take it with a grain of salt, but that’s the order of magnitude.)
To temporarily overcome this, I hooked up my little BS170 board from the Si5351-Amplifier to the output of the K2ZIA board’s amplifier, which gives me about 4W output at 1 MHz (and decreasing above that). This gave me enough power to test the coupler (and the K6BEZ software), and both seem to be working fine. I’d like not to build another 5W amplifier just to use as an antenna analyzer, though, so once a shipment of 1n5711’s arrives from eBay, I’m thinking of just replicating K6BEZ’s detector circuit, using a resistive bridge and op-amps.
Hear you on the air!