larry@kitty.UUCP (Larry Lippman) (12/13/88)
In article <586@aoa.UUCP>, carl@aoa.UUCP (Carl Witthoft) writes: > >A friend of a brother of mine is an artist who is interested in two items: > >1. A battery that can be trusted after 100 years of storage, > >2. A clock of some sort to alarm after 100 years. > > any hunk of calibrated radioactive stuff with an attached geiger > counter. When the count rate drops to a preset limit, fire the alarm. The problem, as I see it, is not so much finding a suitable power source, but finding an electronic timing and control circuit which will work for 100 years. This is not a trivial task; it is not a trivial assumption that say, a CMOS timing circuit will work for 100 years without failure. There are various types of failure modes in solid-state circuits which would manifest themselves over this period of time; such failure modes include, but are not limited to: (1) undesireable migration and diffusion effects; (2) corrosion effects; and (3) growth of micro-fine metallic "whiskers". Present reliability design and testing methods for semiconductors do not cover the eventualities of a 100-year service life. A pioneer in the development of components and circuits with a long service life is Bell Laboratories. Some of these interesting problems have been discussed within the "Bell System Technical Journal". A most interesting series of articles appeared between 1955 and 1957; these articles dealt with the problem of designing undersea telephone repeaters for the first Transatlantic Telephone Cable (i.e., the first _voice_ cable, which was not installed until 1956). These undersea repeaters had a design life of something like 40 years, and the BSTJ articles discussed such problems as growth of metallic whiskers on various components. I would recommend these articles as interesting reading. <> Larry Lippman @ Recognition Research Corp., Clarence, New York <> UUCP: {allegra|ames|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry <> VOICE: 716/688-1231 {att|hplabs|mtune|utzoo|uunet}!/ <> FAX: 716/741-9635 {G1,G2,G3 modes} "Have you hugged your cat today?"
dant@mrloog.LA.TEK.COM (Dan Tilque;1893;92-101;) (12/19/88)
Larry Lippman writes: > > The problem, as I see it, is not so much finding a suitable power >source, but finding an electronic timing and control circuit which will >work for 100 years. This is not a trivial task; it is not a trivial >assumption that say, a CMOS timing circuit will work for 100 years without >failure. In a fairly recent electrionics trade mag, one of the columnists was commenting on a DoD spec which required that a part have a MTBF (Mean Time Between Failures) of about 77 years. I think this part was in a field radio or maybe it was a field radio. Perhaps the originator could get together with whatever contractor fulfilled this spec. --- Dan Tilque -- dant@twaddl.LA.TEK.COM "Computers are useless. They can only give you answers." -- Pablo Picasso