local demolition of SAGE building

[Mike Loewen]

On Tue, 19 Jul 2005, cswiger wrote:

> I understand. They must have run the 49,000 tubes pretty
> conservatively to minimize failures (Unlike what we always
> hear about ENIAC).  Even in 1976 my Heathkit VTVM got
> solid state plug in FET replacements for the two tubes ;)

    Preventive maintenance played a huge part in the reliability of the 
system.  The Q7 had a marginal checking system, which was used to test 
sections of the computer by slightly increasing or decreasing voltages to 
the modules.  By running diagnostics with margins, we usually got a heads 
up on which tubes were about to fail, and could replace them before they 
actually failed in use.


Mike Loewen				mloewen at cpumagic.scol.pa.us
Old Technology	http://ripsaw.cac.psu.edu/~mloewen/Oldtech/