Didn't the small amount of memory they had just completely stop working? I'd love to see or hear how they managed to reprogram this damn thing and make it work again. Amazing.
- It did die- one specific chip carrying the code for packaging the data for transmittal died. They kind of described it in this article. So they are now partitioning out the memory left in other parts of the computer system and copying small portions of the packaging code into those other memory blocks so it can still be successfully run.
  The fact that any electronic component on this probe still works is just freaking wild.
  
  I'm curious as to what would make them stop working anyway. It's in space; no air, no moisture, no bugs, no dirt... Unless it hit something/something hit it, I don't understand how things would degrade in it unless it's just the expansion and contraction of things as they heat and cool from the electricity running through it. But wouldn't that take way, way longer? Is it just the components used at the time not being very good for longevity? It's not using vacuum tubes or things of that nature is it?
  
  Semiconductors are subject to electron migration- in normal operation, silicon is doped to be either P (missing some electrons) or N (extra electrons). It can eventually break down the doping enough such that a semiconductor no longer "switches".
  Similarly, unshielded gamma radiation from the sun and space in general is pretty rough on solid state electronics. It is ionizing radiation so it can affect some semiconductors over time as well as actual mechanical changes in some metals.
  Also consider that the Voyager probes rely on a RTG (Radioisotope Thermoelectric Generator) for their electrical power, a device that uses nuclear decay heat to generate power. These sources also slowly lose output over time as the radioisotope is consumed and the thermoelectric couplers break down due to the radiation exposure. When Vpyager 1 launched, it produced 470 watts of electrical power for its sensor and computer systems- around 1998 they started turning systems off, and by now it's down to around 200 watts, which is not much considering the power demands of the communication dish. When supply voltage starts to drop, previously OK electronics can begin to show errors where the defects exist.
  
  Extreme temps or solar radiation.
  
  Cosmic rays. Wikipedia entry
  
  Even in absolutely dark space, there are kinds of things around. A tiny spec of dust, a large fragmented rock, radiation...anything. this was a largely unshielded device as we know it today. It had some foil to reflect radiation from the sun, and I think that's about it. Even in space, everything will degrade over time from one thing or another given enough time. This thing has had 45+ years.
  I saw this documentary once where a spaceship picked up one of them and it was crazy nuts.
  
  All modern integrated circuits have a limited life expectancy to them because of the natural decay of the photolithographically created microstructures and doping of the silicon crystal. The main driver behind this progressive loss of functionality is electromigration. For larger feature sizes, which translates to older and less capable technology, electromigration is a negligible effect, but it is particularly bad for high performance modern CPUs, especially those smaller than the 200 nm node. While Voyager's ICs are primitive, they are also old and it is very expected that they fail after several decades.
  
  That's so crazy, I love this kind of stuff

27 comments