Follow

more fun with old electronics: i finally fixed this radio from 1967. i should have figured this out faster, but it's a weird bug. something was pulling down the base of one of the output transistors, but the transistor itself was fine, and there was nothing in circuit that could've done it. it was the heatsink. the transistor had a low resistance connection between the base and the case that should not exist. the reason is interesting: (1/2)

these 1960s germanium transistors have a tendency to grow "tin hairs". yes, you read that right. NASA has quite extensive coverage (this bug also crashed satellites): nepp.nasa.gov/whisker/ (here's a short summary: nepp.nasa.gov/whisker/anecdote) once i figured it out, i got a new old stock "matched pair" of output transistors. these are from hungary and don't really suffer from this bug, since...uh...socialism. the casings contain less tin, because there was a shortage. 😁 (2/2)

this is a really interesting bug, because it led (and still leads) to widespread & costly failures, among other things several nuclear power plants and at least 8 satellites: nepp.nasa.gov/whisker/failures
and yet, hardly anybody has ever heard of it.

@aurora and I heard it's happening again with lead-free soldering (RoHS…)

Sign in to participate in the conversation
mastodon@bau-ha.us

social.bau-ha.us is your friendly neighbourhood Mastodon instance!
Proudly presented by Maschinenraum Weimar .
This is a small instance with a focus on community. Feel free to join, but please be aware that there are some rules (in short: don't be a bunghole. Check the link for our definition of "being a bunghole".).
Primary languages are German and English.
Be excellent to each other! (J.W.v.Goethe, 1932) [[Citation needed]]