I’ve been running my outdoor nRF52 project for a few months now, and I thought I’d share some lessons learned for outdoor enclosures. Thanks to all of the helpful folks on this forum for helping me out.
My setup - I use nRF52 boards to gather different metrics in my garden. So, my boards are inside weatherproof enclosures in direct sun 24/7 in central Texas. We get freezing temps and solar flares for our seasons.
Use quality weather proof enclosures with opaque covers. It’s gonna cost you but it’s worth it. Use clear covers for prototyping to check on moisture.
Monitor chip temperature. It’s a sure indicator of your system is running ok.
Vent. Use high quality vents that keep moisture out and allow air to flow. At a minimum have a vent towards the top of your enclosure. Ideally one on top and bottom is ideal. Inverted bucket can work unless your enclosure is close to the ground (like mine). A good rain and you’re hosed.
Use Corrosion X and conformal coating, but keep your antenna clear of conformal coating. I typically leave the uC uncoated to allow heat to dissipate. Corrosion X is great stuff and I paint the entire inside enclosure with it. I use conformal coating on my exposed solder joints and other circuits.
Battery choice. This one is tough if you want to solar power your project. LiPo charging circuits are the most accessible for 3V uC but also volatile. LiFePO4 is probably the best but I use LiPo since there’s lots of chargers available.
If using solar with LiPo, use a quality solar charger with thermal protection. It’s not cheap, but worth the peace of mind (I haven’t tried my setup in summer yet, so we will see how it holds up).
Be prepared to spend money. Going cheap on any part will likely lead to failure if you’re a novice hardware guy like me.
czu001:
So, my boards are inside weatherproof enclosures in direct sun 24/7 in central Texas. We get freezing temps and solar flares for our seasons.
LOL!
czu001:
7) Be prepared to spend money. Going cheap on any part will likely lead to failure if you’re a novice hardware guy like me.
It's heartening to see this. It's not because you're a novice, it's because cheap is good - until it's not. 97% of the posts here are people trying to do X for a buck two-eighty after they've balked at the price$ for quality gear.
"The bitterness of poor quality remains long after the sweetness of low price is forgotten." - Ben Franklin
dougp:
It's heartening to see this. It's not because you're a novice, it's because cheap is good - until it's not. 97% of the posts here are people trying to do X for a buck two-eighty after they've balked at the price$ for quality gear.
It can sometimes be difficult to distinguish "grossly over priced" from "good value" and, at the other end, to distinguish "cheap" from "good value".
And Arduinos complicate this by making it relatively easy to substitute some of your own time for the high price - and get some pleasure along the way.
You can pay about $400 to $500 for a model railway DCC controller or build the equivalent functionality for about $40 or $50 with an Arduino. For some people the Arduino option would make sense, but for others it would not.
You can pay about $400 to $500 for a model railway DCC controller
Something else I've learned... the price of electronics seems to vary quite a bit for the same functionality. For instance, one PCB manufacturer will charge $.50 per board while another $5.00 for the same board. I've also seen small solar panels with very close to the same power requirements go from $1.95 to $7.00!
Of course some of the electronic components are worth the money (if reasonable) while others not so much. I'm not sure WHY there's such a variance, but it's very difficult to sift out the "good value", "cheap", and "grossly over priced" without extensive testing or experience (IMO).
For instance, the $.50 PCB manufacturer has worked out great! But.. maybe not in mass production? I dunno about that.