I left 2 devices in their enclosures out in the sun during the day and it was hot enough to deform my PLA enclosures. I pulled them back in out of the hot sun and I was happy to see that the devices were still working, but over the next week I noticed that the devices started to loose power and come right back on.
It looks like someone is pulling the power and immediately reconnecting it. Perhaps the day in the sun caused permanent damage? This happened on 2 devices and both show this intermittent restarting. Once every 10 -20 mins
I was curious if anyone else had experienced an issue like this from the sun’s heating. I’m not next to the devices and would be interested if anyone can verify if something like this has happened to their devices. I’m trying to see if there are any similar anecdotes of something like this occuring?
The PMIC is noted for running warm. It could well have been damaged in the heat. One way to test could be to use a lab power supply and put in 3.8VDC after the PMIC.
Changing the PMIC out is not easy. You’ll need a hot air rework station if this is what is wrong.
I have a number of devices that are mounted in full sun for months / years on end. I was initially concerned about this issue so I put a temperature sensor on the carrier board for my Electron. I have yet to have an Electron fail even though it gets quite hot in those little boxes.
I have attached the temp graphs for a couple exposed sensors (many of mine are in the woods and don’t see direct sun). I think the biggest issue is not the extreme temperatures (at least here in Raleigh) but the temperature variation which is 90 degrees from winter lows to summer highs. If the seals are too tight on my boxes, the doors can occasionally be blown open on a hot day.
If you’re referring to the equalization valve, I’m not so sure about that. I use Pelican brand cases as well and the new valve appears to be more of a non-return valve. What I’m referring to is a different device with greater surface area.
I tested it a couple days ago. It is not the same thing. It is not a sealed membrane. The membrane may be PTFE but it is not a sealed membrane, and likely slowly leaks wet air in when a hot enclosure cools and pulls vacuum.
The Pelican valve is a solution to altitude changes sticking the lid shut. It is not a hermetic seal.
I was only trying to help you with what an industry standard solution is to this problem. Gore sells a lot of these to enclosure manufacturers for a reason.
Wow! Thanks everyone for all the data! I love this community!
I have the did some more digging and have found that Intermittnet Restarting only started with a recent software flash. SO… based on the feedback from @chipmc@RWB and @LabSpokane I think its very likely that this is a software problem.
Today, I flashed the older version of the code and the intermittent resets went away. I’ll be sure to update more if we find thermal damage, but it seems like it was not a problem due to heat.
I'm with you on the fact that it does not keep water vapor from entering the case regardless of what they claim or use as a membrane.
@Adam42 Good to hear you have found the real issue causing you problems.
There have been reports of disconnects on the 0.7.0 firmware so chime in on that thread if that's the version your running also since they are trying to figure out what is going on.
Well, I have observed this on a couple occasions. I am using the Bud Industries NBF-32302 case. I have found these open with the zip tie seals still in place. It got me thinking about the temperature changes.
If you allow me to break out the Ideal gas law (remember PV=nRT back from school) and you make a couple big assumptions: 1) That you sealed the box on a very cold day and 2) That it is air tight then, over the range of temps I shared above you could have a 0.23 atmosphere pressure difference or 3.38 psi. The lid of the box has an area of ~20 square inches so there could be a force of 60 pounds (if I did my math right). These numbers are rough and I am not sure if this is enough to force the door to pop open but, it does not sound too unreasonable.
Perhaps the Goretex idea from @LabSpokane is worth looking into.
