MOVs are interesting critters whose primary attribute is very low cost. The
1.5 joule, 3 joule and even 10-15 joule devices cost pennies each.
In glancing over the correspondence here, one thing I haven't seen (unless I
missed it) is a discussion of the fact that MOVs are doped grains of zinc
oxide arranged in series to build up to the breakdown (and operating)
voltage required. Each grain breaks down at a very low voltage, like 2v, so
to operate across a 240v line it takes more than 120 grains in series.
Their failure mechanisms are actually quite well understood. We understood
them very well at G.E. in Syracuse, NY back in the mid-1970s, which is ~30
years ago now, and nothing's really changed. What happens during a
transient event is each grain breaks down, draws increased current and
dissipates power as heat. If the semiconductor material gets too hot it
changes in resistivity permanently (lower) and from that point forward will
break down at a lower voltage. A 240v "operating" (and 360v-480v breakdown)
MOV can become a "180v operating" MOV pretty quickly, and draw such
excessive line current it simply fails altogether. Failure due to prolonged
overheating is a low-resistance fault, nearly a short circuit. Should blow
a fuse or breaker somewhere.
However, failure due to a very high surge energy which causes the
terminations to blow right off the semiconductor material is usually an
"open," which of course means now you have zero protection since all you
have left across the line is a pair of disconnected wires.
There is another insidious mechanism, though: Very slow overheating and the
time to failure that requires. When the grains start breaking down at lower
voltage and drawing more current, unless there's a new transient event to
make them breakdown for good, there's a pretty fair chance the MOVs
themselves will simply get hotter and hotter and not blow a fuse or breaker
because the leakage current isn't high enough. This eventually avalanches
into thermal collapse and the short-circuit failure we all hope for, but it
can occur very slowly -- even over a matter of hours. During that time, the
devices can become scalding hot without failing. Although the devices
themselves are not flammable, or are flame-retardant, and should be
self-extinguishing to achieve their UL approval (which they do have), the
materials nearby may not be quite so tolerant of flame or temperature. We
had many examples in Syracuse of MOVs burning up nearby combustibles, and
From: Jim Jarvis [mailto:firstname.lastname@example.org]
Sent: Wednesday, August 03, 2005 2:59 PM
Subject: [BULK] - [TowerTalk] MOV's
I did some consulting work for a manufacturer of surge
protection equipment, several years ago.
We evaluated MOV's and their failure mechanisms, using
Keytek surge generators (whom I represented at the time).
Whether we WANT them to fail shorted or not, they fail
either shorted OR open, depending upon the total number
of Joules they have absorbed, the direction of the currents
which flowed, and what the latest peak does to them. They
do it without asking us permission.
My conclusion at the end of the project is that an uncontrolled
variable in the MOV production makes it impossible to predict in
advance whether they are predisposed to fail in one mode or another.
Simply exciting them differently did NOT produce different results.
Yes, MOV's operationally clip peaks. They just don't do it forever,
and they have no way of telling you that they just died, unless you
get lucky and they short and blow a fuse. That's maybe half the time.
ANSI/IEEE C62.11 sets the standard. There's another for the
energy waveform of the test excitation, which I forget, and
google isn't yielding quickly.
With respect to my personal protection, I stand by my statement:
An MOV based solution is at best a short term solution. You'll
never know when 'short term' is over, until your equipment is fried.
Hence, I believe it's no solution at all, unless you're selling cheap
consumer crap to the unwary/uneducated.
In which case, the variable of interest is return on investment, not Joules
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless
Weather Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any
questions and ask for Sherman, W2FLA.
TowerTalk mailing list
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless Weather
Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any questions
and ask for Sherman, W2FLA.
TowerTalk mailing list