I like the comment about vacuum tubes – yeah, they ran hot, they had to (that thermionic emission thing). And for transmit tubes (like the 6146B), not only did they run hot, the voltage on the exposed anode connector could/would/wants to kill you.
Yet these designs were manufactured to be reliable, and they were and still are. In the ham radio area, the Collins S-line of receivers, transmitters, transceivers, and power amplifiers are still prized today – and they’re filled with vacuum tubes. They were designed to keep the hot stuff (tubes) on one side of the chassis, and a lot of the other parts away from the hot stuff.
Our modern stuff I believe is more reliable (at least the hardware side of things – software is a different story entirely).
One reason is our reliance on surface-mount electronics, and the reflow soldering process.
That reflow process preheats the board and all its components to between 150 and 180 C and soaks them in that 150 - 180 C range for a few MINUTES. The actual reflow ramps the temperature up (over a minute or more, a change of 3 C per second max is the usual spec) to around 230 C for 10 to 30 seconds or so to get the solder paste to reflow, and then cools down (controlled cooling, usually 3 C per second max).
I would not like to take most of the components of a Collins S-line box through the modern preheat and soak temperature cycle! Most of the components won’t survive! They weren’t designed to!
If we look at a Raspberry Pi based ADS-B system, the Pi and the SDR are both surface mount boards – they’ve gone through the oven, quite literally. All their parts are designed to survive reflow assembly temperatures. Depending on what curves you use, that’s 6 to 8 minutes or more above 150 C. The total time spent above 100 C can be much longer, possibly 10 minutes or more.
Using a quality (micro) SD card is important – I’ve not had problems with legitimate Samsung memory. (And there are things that you can do in software, like run certain parts of the system in tmp directories that stay in RAM to reduce wear and tear on the SD card – and newer versions of the PiAware software use all these tricks).
But the core hardware has been through the oven. Even the PiAware filters are surface-mount and reflow inside (well, the connectors look like hand solder after the reflow process).
After going through that surface mount assembly process, running at 60 - 80 C isn’t a big deal.
(You can look up components on sites such as Mouser and Digi-Key and see what their manufacturers recommend as far as reflow temperature/time profiles. You can also look up reflow soldering. Adafruit has some great videos of their automated assembly lines, from applying solder paste to boards to pick and place of components, and through the reflow oven. There are industry standards, but as as been oft stated, the great thing about standards is there are so many of them.)
(I picked up a great t-shirt that aptly summarizes the state of the art in software. https://www.amazon.com/gp/product/B01ALNJ6GY – please take a look, and congratulate obj and the PiAware dev crew for the efforts they put into building stable,reliable code. It ain’t easy!)