Today marks a major milestone of progress on the J-Box component of the software-defined antenna project. After many months of hardware and firmware development, we now have a fully functioning prototype!
It’s been a long haul, with many challenges, but despite them all there is now light at the end of the tunnel – and it’s definitely not an oncoming train – it’s a self-contained RF switch known as a J-Box. In version 0.14 of the device, the following observations were made during testing.
- Frequency counter and latching relay switching worked perfectly, based upon the DIP switch ham band cutoff frequency configured.
- Frequency counter functioned perfectly with RF harvest / pickup wire from 1 watt through 100 watts.
- RF Harvest circuit produced nominally 9 volts rectified DC at 5 to 10 watts transmit RF, spiking to 10.5 to 11 volts after RF transmit stops (same as earlier observations).
- RF Harvest circuit produced maximum of 12 volts DC at 100 watts of transmit RF (it is clamped by a zener diode).
- Firmware entered Ultra Low Power sleep mode correctly upon reboot and at the end of each RF transmit / switching cycle, and awoke with the arrival of new RF
- RF signal must be maintained for at least 4 seconds to trigger a band switching cycle. A bug was noted whereby shorter transmissions were correctly read, but did not reset quickly enough to sleep state when brief RF (e.g., CW) is encountered. This bug will be resolved in the next firmware update.
Testing was performed using a centerfed 20 meter dipole in my kitchen (SWR was a bit high, but it was raining and I wanted to complete the testing today). The rig used was an ICOM 7300.
The hardware design has evolved, as well. Below you will find images of the current prototype, along with an explanation.
In the above photo, we see the J-Box PCB, along with a ribbon cable to the USB serial interface (Prolific PL2303). The two screws protruding upward from the PCB are where the incoming RF Harvest signal arrives. The DIP switch is used to configure the “cutoff frequency”, the ham band where the device will open/close the relay to add or subtract antenna elements. The rest of the components are shown in the schematics below.
Above we see the rear side of the J-Box PCB, where the surface mount components and wireless MCU board are attached. I ran out of SMC zener diodes, so there’s also a messy patch or two as well (leading to next hardware rev. 0.15). D1 and C2 have been removed as not required.
Below you will find the latest schematics and firmware code as of today. Work will continue progress. Getting very excited to finally have a working unit!
Download the Rev0.14 firmware below (rich text file can be viewed with Word)