A long but hopefully interesting story – I need some help to save the world (or at least one small, precious part of it) – read on!
Background: I’m a Park Ranger at Moreton Island National Park, Queensland, Australia.
My main job on this island is controlling (read “killing”) feral pigs. It’s been my focus for about 5 years, and I’ve managed to take out nearly 300 of them. I hunt, use poison and traps. I’m down to the last few dozen – so that means getting “smarter” with my trapping. The island is ~150 square kilometers and getting around is dependent on tides, swell and time.
Last year I built a prototype remote trigger for one of my traps. Simple stuff - 12v solar system, weather proof enclosure, Electron, INA219 to monitor the SoC and a relay to put 12v into a car door lock actuator to release a trebuchet trigger and drop a gate shut on a trap. The traps are steel mesh, about 10 meters wide with a wooden guillotine door. I use a dear feeder (solar powered on a timer) to toss some corn around and monitor the proceeding from afar via 3G connected trail cameras (from Spartan Cameras: http://www.spartancamera.com.au/gocam-camera/ ).
I have 10 traps scattered around the island. When they turn up at an area, I feed the pigs up over a week or so, wait till they are all comfortable with going in and out of the trap then watch the trap from the comfort of my home and trigger the trap via an IFTT applet. Works great – apart from being woken up by my phone at 3am to “work”.
Here’s the first successful test (I’ve obscured the 2 victims after dispatching them).
…in they come…
…drop the door (they are at the back of the trap)…
…and I have a 2 hour drive to deal with them.
Trap.
3G camera – motion activated – home-made 12v solar supply.
Deer feeder on the right.
Car door lock actuator – trebuchet trigger – the door weighs about 15kg.
Control box, solar panel, magnetic reed switch on the door (in case the camera goes offline) – still using an omnidirectional antennae at this point.
12v battery, solar regulator, and Electron.
When I saw the Generation 3 gear coming through I had grand ideas of switching to Borons to run the trigger and putting a xenon in the feeders so I could a) alter the feed time remotely (No point dropping corn of the ground twice a day when there are no pigs around right?); b) use an ultrasonic sensor to monitor the level of corn in the feeder; and even c) automate the triggering of the trap (PIR and ultrasonics).
So, I now have a box full of Borons and Xenons and one Argon (for testing locally) and I’ve been making “magic happen”.
My problem - Cellular reception. This island is 30 – 50km from any cell tower (apart from a few small, local towers serving the towns). It’s the 3rd largest sand island in the world, has interesting topography and a lot of other challenges. My trap/camera locations are subject to getting cell signal – 1 to 2 bars typically – from near the top of a strategically placed tree. Those towers big enough to reach here are (almost) exclusively from Telstra.
So, I have 2 Electrons. One in my first remote trigger and the other in my home at Cape Moreton monitoring the power system.
The Cape Moreton Lighthouse - 41km from the tower I get service from.
(I live in the house to the right of the lighthouse).
Both of these Electrons have been happily working on Signal strengths around 28 – 32 and signal quality of around 50 for about a year. To improve the reception of the trap trigger I use a directional 3G antenna pointed south toward a cell tower on Stradbroke Island – it doesn’t miss a beat.
Sate of Charge graph – IFTTT from Particle to Google Sheets – scripts keep the last X days of data.
The Borons however are basically useless here. When they (eventually) connect to the net they report the same signal and quality as the Electrons (30-ish strength and 50-ish quality) – but they constantly drop out (fast blink green – for hours sometimes – 8 hours during the day is not uncommon). Even with a directional antenna (giving signal strength of 60 -70) the Borons drop out then go into a spiral of disconnects… then spring back to life hours later. All at the same time, the Electrons just sit happily doing their job (quietly mocking me for trying to embrace the mesh). If I take the Borons to the mainland and use the standard antennae they work fine (reporting a signal strength of around 60 – 70).
So:
• What has changed in the change from Electron to Boron as far as the Cellular Connection goes?
• I suspect my Borons are trying to talk to non-Telstra carriers – how do I discourage that?
• How can I troubleshoot all this? (Given the absolute crappy reception (and I only stay on the mainland for 1 or 2 days a fortnight) - I do indeed own a Particle Debugger! Can this help me?).
I don’t want to have to use a Telstra SIM. I use them in my 10 cameras, and they cost me a $150 a year, each – and with the cameras (despite taking 100s of HD pictures) I’d be lucky to use 300MB per month per camera -so much wasted data.
So - your thoughts?