mark@mips.UUCP (07/07/86)
Isn't it common design practice to place surge protection devices
(e.g. GE MOV-Varistors) __after__ the 110VAC fuse? If so,
the many-ampere surge current would flow through one blade of the
AC plug, through the fuse, through the varistor, and out the
other blade of the plug. Thus, the fuse "protects" the varistor
(and the equipment!) from subsequent surges. It also gives you
the opportunity to inspect the varistor when you replace the fuse.
Why not extend the technique to an n-stage (n>1) series-shunt
ladder network in which the series elements are fuses and the shunt
elements are MOV's?? Only the fuse nearest the AC line need be
the "rated" current of the equipment; the rest of the fuses could
be, say, 150% of the rated value. With this technique, the current
surge is shared (perhaps unequally!) among n varistors, so the peak
current seen by each is (approx!) 1/nth as big, so the expected life
time of each varistor is about n times longer. Also, the equipment
remains protected even if some of the varistors fail, as long as one
or more is still alive. This would seem to extend the mean-time-
-before-complete-varistor-failure by another factor of n.
--
-Mark Johnson
UUCP: {decvax,ucbvax,ihnp4}!decwrl!mips!mark
DDD: 408-720-1700
USPS: MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086