– Copied from customer request –
Are you aware if we’ll need an enclosure for stratospheric conditions for deploying SwitchBlox Rugged + Rugged SOM in the stratosphere?
To operate in the stratosphere, you are very likely going to need some kind of enclosure for SwitchBlox Rugged + Rugged SOM, but it doesn’t need necessarily it’s own separate enclosure. I assume you have some kind of overall enclosure for your device?
If you are skillful in how you integrate the bare boards into that enclosure, SwitchBlox Rugged + Rugged SOM can operate at the low temperature and low pressure conditions in the stratosphere.
Rugged SOM and SwitchBlox Rugged can operate down to -40°C, and we have other customers operating at this temperature. Low pressure generally results in less convective cooling, meaning chips can overheat more easily. This actually works in your favour in stratospheric conditions, because the chip’s self-heating often prevents the silicon junction temperature from dropping below the -40°C. Direct sunlight, however, can cause fast heating and reverse this benefit, so it is important the boards are not in direct sunlight.
Regarding radiation, if your enclosure is at least 2mm aluminium or equivalent, you will benefit from some basic alpha radiation blocking and UV blocking.
For vibration, generally the Molex PicoBlade and PicoClasp connectors operate well under vibration however some customers do prefer a more rugged connector for such applications. That generally means a larger board size, and we’re aware of some customers are adding epoxy to the ports on our boards.
We haven’t ever had a report of a connector shaking loose despite over 100,000s of boards using that connector out in the field*. We’ve also tested it here (see page 13, different board but same connector.)
*Not having a report of it doesn’t mean it’s not happening, but I’m trying to add some empirical context here. Generally our customers are very vocal when things don’t work (and we actually really appreciate that). We’re responsible for probably millions of Molex PicoBlade/PicoClasps out there in the wild in real world applications, and I can genuinely say I’ve possibly only had one report of a connector coming loose (I can’t even remember actually). If this was a common or pernicious issue, you would expect us to be inunndated with reports of connectors coming loose, and if that was the case, we’d be damn sure to move to a different connector system!
Being totally upfront the most common fails related to the connectors are:
a) Connector gets ripped off because the engineer pulled too hard in a horizontal direction (or because of inadequate strain relief in application.
b) Poorly crimped cable, or overuse of the cable (after about 50 mate/demate cycles, the crimp inside the cable starts to widen, resulting in less reliable contact. Usually that means you gotta replace the cable, and this doesn’t happen when the cable is just solidly plugged in).