Surge "suppressors" revisited: a letter to Larry
Andy Baird

At last month's Princeton Macintosh Users' Group (PMUG) meeting, 
a member got up to tell of his network problems. He acts as 
administrator for a large network of Macs and VAXes, and although 
the machines were supposedly being protected by fairly expensive 
surge suppressors, he was experiencing repeated and costly serial-port 
failures on the Macs, apparently due to surges. Some machines 
had had their serial interface circuitry replaced as many as 
five times! The symptoms were consistent with those of a serial 
device damaged by a surge protector dumping energy into the 
ground reference line.

Reading about Zero Surge in the June PMUG Newsletter, he ordered 
half a dozen Surge Eliminator units and installed them on the 
Macs which had been the hardest hit. Subsequent surges have 
damaged other machines in the same area, but not one of the 
machines protected by a Zero Surge device has been damaged. 
He and others have suggested that it might be a good idea to 
run some information on Zero Surge, which I covered in detail 
a year ago. Meanwhile, another friend of mine has been having 
trouble convincing his company to buy Zero Surge devices instead 
of conventional surge suppressors. So here's a brief summary 
of the reasons I personally recommend their product...

Larry--
You asked for justifications you could present to your Purchasing 
Department, who are questioning your requisition for Zero Surge 
Inc.'s Surge Eliminators. Well, I don't know whether this will 
change their minds, but here are the reasons as I understand 
them.
For what it's worth, there are two main reasons to specify Zero 
Surge. First, the vast majority of conventional surge suppressors 
use metal oxide varistors (MOVs). Although these surge suppressors 
themselves may cost quite a lot, the MOVs are fifteen-cent parts 
which wear out in normal use. In as little as six months to 
a year, the short spikes which occur several times a year in 
a normal operating environment will reduce and finally destroy 
an MOV's ability to do its job. (GE's MOV handbook, among others, 
amply documents this.)
The kicker is that although the surge suppressor unit may have 
fancy "protection on" indicator lights, the MOVs can be completely 
worthless and the lights still light up! The indicators say 
you have protection, but you don't. And there's no way to test 
MOVs without a few thousand bucks worth of equipment.
I've examined units such as the Kensington Masterpiece in which 
the MOVs were completely blown apart, and the plastic case partly 
melted by the failure, while the front-panel indicators said 
nothing was wrong! In another case, I physically cut the MOVs 
out of a $70 surge suppressor--without affecting its indicator 
lights at all. What's more, when a sizable surge hits, the worn-out 
MOV can explode or catch fire (I've personally seen both), so 
if you have a surge protector in a plastic case, it's a fire 
hazard.
Zero Surge uses a completely different circuit principle, with 
no components that are stressed or will self-destruct. Doesn't 
matter how many surges hit the Surge Eliminator box or how big 
they are--the Zero Surge unit won't wear out, and it will never 
pass a surge. The shortest rated life of any of its components 
is 26 years at normal temperatures.
The second reason not to use conventional suppressors is perhaps 
even more serious. Conventional units get rid of the surge energy 
by dumping it onto the ground line. This is like trying to sweep 
it under the rug--it isn't really eliminated; it's just moved 
somewhere else. Trouble is, that ground line is also the signal 
ground reference for LAN connections, modems and other devices 
connected to your computer. Dump a 3,000V spike onto that line, 
and all of a sudden your modem is seeing +5V on its signal line 
and +3,000V on its ground--a recipe for a dead modem. Again, 
this is not just theory; I've seen it happen.
Zero Surge's Surge Eliminator does not contaminate the ground 
line; instead, it stores the surge energy in a cascaded series 
of large capacitors and then releases it s-l-o-w-l-y, dribbling 
it back into the neutral line at a rate so gradual that it can 
do no harm. Thus, your modems and LAN connections are never 
exposed to the surge energy.
There's a mounting body of engineering reports and other evidence 
indicating that Zero Surge's operating theory is correct, and 
that the dangers I just described with respect to conventional 
surge suppressors are real. Well-known LAN vendor Novell, among 
others, recommends Zero Surge to its customers.
If the folks making this decision were to attend the upcoming 
Surge Protection Forum, sponsored by the National Institute 
of Standards and Technology in Washington, they would hear a 
number of papers presented which substantiate Zero Surge's claims. 
Franois Martzloff, head of NIST's research group on power anomalies, 
has published several papers which corroborate Zero Surge's 
operating principles based on his group's research.
I'm sorry this explanation had to be so long. Perhaps one of 
the reasons your buyers are distrustful of Zero Surge is simple 
lack of knowledge. While an MOV is as easy to understand as 
a fuse, Zero Surge uses a more sophisticated approach. If that 
is the root of their objection, it's equivalent to arguing that 
these newfangled pocket calculators are too complicated and 
we should all just stick to our slide rules. I don't know what 
you can do to get them to understand the facts. Maybe your new 
Technology Coordinator can help educate your Purchasing Department.
It's hard to say what to recommend to you, Larry. The $70 surge 
suppressors they want you to use (I've taken them apart) have 
three fifteen-cent MOVs and a handful of passive components. 
If you get one, it will protect your computer--somewhat--for 
six months to a year; until the MOVs wear out. Of course, you 
will have no way of knowing when that happens. It will go on 
cheerfully indicating you have full protection, even if its 
MOVs have been physically blown apart by a surge.
When it is working, it will be dumping surges onto the ground 
line, endangering your LAN, your modem and other devices connected 
to your computer. On balance, I'd say that if you can't get 
a Surge Eliminator, you may be better off running without a 
conventional surge "protection" device.  --Andy
This article is reprinted from DIALOG, the newsletter of the 
Princeton Macintosh Users' Group, and may be reprinted in substantially 
unedited form by other noncommercial groups. (Special permission 
has also been granted to Zero Surge Inc. to and distribute this 
article.)
