I've seen the butt weld (or crimp) of the wire in hermetic units fail due to corrosion in both HP and Tek gear. In all cases (about 6), the gear appeared to not have been protected too well from the elements, or was exposed to high humidity for a long time. The good news is that it fails open and is pretty obvious.
Dave ----- Original Message ----- From: "Chuck Harris" <cfhar...@erols.com> To: "Discussion of precise time and frequency measurement" <time-nuts@febo.com> Sent: Monday, February 8, 2010 6:46:54 AM GMT -08:00 US/Canada Pacific Subject: Re: [time-nuts] The Smell of Tantalum in the Morning The history of tantalum failures is wide and varied, but there are some common characteristics: 1) The tantalum is in a power supply circuit and receives a rapid ramp from 0V to operating voltage. 2) The tantalum is spec'd close to its operating voltage, very close.... 5V on a 6.3V part, 12.5V on a 15V part... 3) The tantalum is dry slug, and is sealed with epoxy. 4) The instrument has been powered down for an extended period. HP equipment from the 1980's is pretty immune to the problem because they typically use hermetically sealed mil spec tantalum capacitors. Tektronix equipment from the 1980's is infested with tantalum problems because they used the cheap epoxy dipped parts. Tantalum failures are pretty rare in equipment that is run continuously. Tantalum has a self healing feature that corrects any small problems while in operating... Large problems result in detonation. Dipped tantalum capacitors of any age are prone to failure. The tendency can be mitigated largely by never allowing a tantalum capacitor to see voltage above 50% of its rating. And finally, powering a tantalum in reverse, will cause instant and irreparable damage. -Chuck Harris Tom Van Baak wrote: > I powered up a 5071A to watch the end of Loran-C today > and was greeted by the special smell that only a mother > board could love. > > Does anyone know the history of tantalum capacitor > failures in ten-year old [HP/Agilent] test equipment? > This is not my first. Last one was more like July 4th. > > Thanks, > /tvb > > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.