Zero Surge model ZS1800 Surge Eliminator 
Product Review and Report 
 
Vernon L. Chi 
Microelectronic Systems Laboratory 
UNC Chapel Hill 
31 July 1990 
 
The views expressed herein are those of the author and do not 
represent the position, official or unofficial, of the University of 
North Carolina. 
 
My first exposure specifically to transient protection was upon 
moving to North Carolina where our Department experienced chronic 
equipment damage from thunderstorms.  Every spring, we would 
purchase dozens of 1488s and 1499s (RS232 line driver and receiver 
ICs) and routinely used them up during the season.  Power supplies 
were a less frequent, but equally sacrificial component to the 
lightning god.  In one rather extreme case, a near strike induced 
extensive internal damage to one of the Department's VAX11/780s in 
which a circuit board plugged into the backplane actually was in 
flames. 
 
So when the Department was funded to erect a new Computer Science 
building in the mid 80's, I took on the task of ensuring our 
equipment and data would be safe from damaging electrical tran- 
sients. In the process I learned a great deal about the arcane 
science of lightning and high energy transient control in general.  
The Sitterson Hall (as our building is officially known) protection 
is probably quite over designed, but as Dr. Martin Uman of The Uni- 
versity of Florida, a well respected authority in lightning research 
advised me, most people who haven't over designed in this field end 
up wishing they had.  The net result is we have experienced no inci- 
dents of equipment damage or even identifiable data corruption due 
to transients since we first occupied Sitterson Hall in 1987. 
 
I first heard of Zero Surge's product in a news posting by Chris 
Bannister to a Macintosh users' newsgroup.  The right words were 
being said, but I was initially skeptical because I had become 
inured to the extravagant claims which unfortunately seem to 
proliferate in the point-of-use surge protector marketplace.  My 
engineering experience had focused on commercial/industrial site- 
wide surge protection, where entire buildings or complexes require 
treatment.  I learned that even in this high-end market, the 
literature is replete with unsubstantiated claims; moreover, I 
discovered there is precious little agreement among respected 
authorities on many aspects of lightning protection.  I did discern 
a few points of generally common agreement, however. 
 
First, when it comes to Acts of God, all bets are ultimately off -- 
you can't guarantee anything you will do will be sufficient in the 
"worst case".  The energy of a single lightning strike can vary over 
an extraordinarily wide range.  Currents can be as low as a few 
thousand amperes but can exceed hundreds of thousands of amperes.  
Durations can be from a few hundredths of a second to over a second. 
The transient can enter your equipment by induction or by a direct 
hit to a conductor (power or signal) connected to your equipment.  
No protector can keep you safe from a sufficiently energetic close 
proximity direct strike.  The best you can do is to design your 
protection to deal with energy levels consistent with a 
statistically acceptable risk.  Unfortunately, the statistics vary 
with general location, weather patterns, and local geography, and 
you won't find any ready answers in the literature which really 
apply to your specific case, although several published "typical" 
line disturbance probability graphs are widely cited and used for 
design purposes.  I note, in this connection, that there are many 
other sources of line transients in addition to lightning, but they 
are generally far less destructive; the design charts I mentioned 
include line transients irrespective of their sources. 
 
Second, your grounding treatment is critical to the performance of 
any protection scheme.  "Ground" is merely an abstraction -- a state 
of mind about your environment. Near a lightning strike, "ground" is 
not electrically grounded at all.  In fact, the NFPA code recognizes 
a so-called "step voltage" which occurs between two points on the 
ground spaced the distance of a typical human step.  This can be 
many thousands of volts.  Simply shocking!  For this reason, the 
NFPA recommendation on personal safety if you're caught outside 
during a thunderstorm and sense a strike is about to occur is DO NOT 
lie down.  The voltage between where your head and feet touch the 
ground can kill.  Instead, place your feet close together and 
"hunker" down as low as you can.  The point I am trying to make is 
that many electronic and computing systems are connected to 
different grounds associated with different wires such as the power 
line, and the signal connections to printers, modems, etc.  If large 
voltages are induced between these various grounds, the potential 
exists for severe damage to the equipment involved.  Therefore any 
protection scheme should strive to minimize the induced voltage 
difference between grounds connected to your equipment. 
 
BRIEF TECHNICAL DISCUSSION: 
 
The basic problem is stated, and two fundamental types of protectors 
are described.  Some of the pitfalls in the design of protectors are 
mentioned, particularly in the selection of components used in them. 
 
Associated with any transient is an amount of charge delivered 
(equivalent to the number of electrons) and an amount of energy 
(proportional to both the charge and the voltage at which the charge 
is delivered).  Physicists tell us that neither charge nor energy 
can be created or destroyed; they can merely be moved around, and 
that energy can be converted into a form we call heat.  So if a 
transient is delivered from a wire to a piece of equipment, the 
charge and the energy both end up inside the equipment where, if 
they are not dealt with, they may wreak havoc.  There are 
fundamentally two kinds of surge protection devices intended to 
prevent transient charge and energy from entering sensitive 
equipment.  These are the shunt type, and the hybrid series-shunt 
type. 
 
The shunt type is simply connected to the power line at or near the 
same place the sensitive equipment is connected.  Its principle of 
operation is to present an alternative easier path for the transient 
charge to take than into the protected equipment, thereby enticing 
the charge to enter the surge protector instead.  Once in the 
protector, the charge is shunted off to neutral (the power return 
wire) or ground, and the energy is typically "dissipated" or 
converted into the form of heat.  The components used for this 
purpose, typically MOVs (Metal Oxide Varistors), Zener diodes, or 
Transzorbs, are the devices which dissipate the energy, and 
consequently they heat up.  Moreover, the MOVs wear out a little bit 
on each hit.  In any case, a sufficiently energetic hit will 
overheat the device and burn it out.  Another problem with the shunt 
type is that the path into the protection device doesn't become 
easier instantaneously.  It takes the components a small but 
critical period of time to get themselves switched on. During this 
brief time, the easier path is into the sensitive equipment you are 
trying to protect, such as your precious computer.  Zap.  It doesn't 
take long to do the damage, and by the time the protection device 
kicks in it may well be too late.  The singular virtue of the shunt 
type protection device is that it's inexpensive.  That's because 
there's almost nothing to them.  Most  if not all "power strip" type 
surge protectors are of the shunt type. 
 
The hybrid type solves the reaction time problem by inserting a 
component between the incoming power line and the protected 
equipment.  This component is usually an inductor, which naturally 
resists any sudden change in the flow of charge through it.  It's 
very much like a revolving door.  If there's a relatively steady 
traffic of people going through the door, it will be revolving at a 
certain rate.  Then, if some impatient person runs at it full tilt,  
--Wham!-- it slows him down to nearly the same speed as the more 
sedate average traffic.  This has the effect of presenting a 
preferentially more difficult path into the protected equipment for 
a brief period, long enough for an associated shunt protection 
component connected to the incoming, or "line" side of the inductor 
to kick in.  Often, a second shunt component is connected to the 
equipment, or "load" side of the inductor for additional protection. 
The Zero Surge protector is the only point-of-use surge protector I 
know of that is of the hybrid type. 
 
The rate of onset of many power line transients is very rapid -- in 
the range of a few nanoseconds (that's billionths of seconds) and 
modern electronic equipment is susceptible to damage in these very 
brief periods.  The behavior of electrical components during these 
extremely brief intervals is often very different than one  might 
expect.  Even heavy gauge wires with very low actual resistance 
exhibit a surprisingly high "resistance" (actually, inductive 
reactance) during this brief time.  Inductors can act momentarily 
like capacitors due to their stray inter-winding capacitance.  
Similarly, capacitors can act like inductors due to their foil wrap 
inductance.  Moreover, the enormous magnitudes of current and 
voltage that can occur during  a transient may cause the
characteristics of some components to temporarily or even 
permanently change.  The inductance of a coil can easily drop to a 
hundredth or less of its normal value under transient high current 
conditions, making it ineffective.  Likewise, the "effective series 
resistance" and residual inductance of capacitors can im4air their 
effectiveness and may even contribute to failure under these 
conditions. 
 
The message here is that components used in protection circuits must 
be understood thoroughly (the engineers refer to modelling the 
components at a sufficiently detailed level of abstraction) and 
selected with care in order to achieve the intended level of 
protection, especially during the very brief but critical onset 
stage of a line transient.  The robustness and longevity of these 
components in service will also strongly depend on their ability to 
deal with large transients without exceeding their ratings. 
 
THE CLAIMS OF ZERO SURGE: 
 
Chris Bannister's electronic news posting about Zero Surge prompted 
me to post a notice locally to inform newcomers to Sitterson Hall 
explaining the level of protection we have here.  The posting 
somehow got back to Chris, which initiated a dialogue -- actually a 
trialogue, as Chris forwarded my comments and questions to Wendell 
Laidley of Zero Surge, and his replies back to me.  Because of the 
pervasiveness of extravagant and unsubstantiated claims in this 
market, I was initially skeptical, and made no bones about it.  
After a few e-mail exchanges, however, I found the technical 
responses to be well considered and basically correct according to 
my experience.  Indeed, after a few more exchanges, I was pretty 
well convinced that  the Zero Surge product had been engineered 
rather than just built. 
 
My curiosity having been aroused, I suggested to Chris that it would 
be interesting to inspect a Zero Surge ZS 1800.  This is in fact 
what happened.  Wendell Laidley agreed to supply me an evaluation 
unit, which I have inspected and have run field trials on.  I do not 
have the equipment necessary to run laboratory tests on surge 
protectors, so I cannot independently verify the coulomb (charge) 
and joule (energy) ratings of the unit, but in field tests on my 
wife's Mac SE, there have been several instances where powerful 
transients have occurred without so much as a wavering of the 
display. 
 
I live in a bad lightning location, and am served by the last pole 
transformer at the end of one of Carolina Power and Light's 
distribution lines.  This is a bad situation, because when 
transients hit the end of a line, it's just like hitting a brick 
wall -- they don't want to stop.  So their effect is magnified.  
When a strong transient hits my house, if any lights are on, it's 
like a photographer's flash tube going off -- you definitely notice 
it.  As I say, there have been three such incidents which my wife's 
Mac has just laughed off.  In the past, we had to make absolutely 
certain that the Mac was switched off if there was a thunderstorm 
anywhere in the area.  Not any more. 
 
My physical inspection of the unit and circuit information supplied 
to me by Wendell Laidley leads me to the following assessment.  The 
unit is well engineered with sufficient attention to detail and 
sufficient understanding of the technical issues that it can be 
expected to perform according to the claims made.  This is not a 
minor point, in light of the snake-oil nature of this market.  It is 
difficult for non-technical users to distinguish between hype and 
fact when it comes to evaluating these claims.  I can assure you 
that Zero Surge's claims are technically well founded. 
 
The ZS 1800 is a hybrid type protector.  I questioned Wendell 
Laidley on the selection of capacitors, and was told they were 
chosen for their low ESR (effective series resistance) and 
inductance.  I queried him about the lifetime, as it's been my 
experience that electrolytic capacitors (the only kind with enough 
capacitance in an acceptably small and inexpensive package) "dry up" 
with age, and lose capacitance.  He informed me that they had 
performed extensive accelerated life testing on these devices, using 
a combination of standard techniques along with their own stress 
tests which repeatedly pulsed the capacitors.  So, even though the 
capacitors look "ordinary" like I'd expect to see in a transistor 
radio, in fact they are premium grade in all respects relevant to 
this application. 
 
I had initially expected a "hash choke" type of inductor as the 
series component of the hybrid.  This would have been a concern 
because it's core would have easily saturated with a concomitant 
loss of inductance for large transients.  In fact, the coil was a 
custom design by Zero Surge, well tailored to the application at 
hand.  It is a large form factor (over an inch diameter by about 
four inches long), single layer coil with a lossy iron core.  This 
design results in very low inter winding capacitance for excellent 
high speed transient blocking characteristics, high saturation 
current so the inductance stands up to large transient currents, and 
dissipative to help control the behavior of the transient in the 
circuit.  I noticed with some amusement that the iron core material 
they had chosen was a bunch of finishing nails!  This would 
ordinarily make a pretty crummy inductor, but for this application, 
the characteristics are well suited. 
 
This brings up an observation about the construction of the ZS 1800 
(as distinct from the engineering).  It's obviously a garage shop 
type product.  Don't get me wrong.  The materials and workmanship 
are just fine.  But a Sony it ain't.  No slick injection molded 
plastic with high tech multilayer circuit boards.  It's very very 
vanilla, with hand-wired (no doubt with pride) connections between 
ordinary power outlets, switches, etc. mounted on a bolt-together 
steel box, and the various components, mostly mounted on a modest 
low-tech circuit board.  This is not high technology, but definitely 
"appropriate tech".  I would predict that if Zero Surge as a company 
survives and prospers, you will see this technology soon displaced 
by "newer tech" packaging more amenable to modern mass production.  
Attendant with this, of course, should be a radical reduction in 
manufacturing cost.  Whether or not this appears as a price 
reduction passed on to the customers remains to be seen.  In any 
event, at this time Zero Surge is in the business of hand-crafted 
units.  Which makes the low price tag even the more impressive. 
 
Finally, a word about the circuit.  Zero Surge has eschewed the 
usual transient energy dissipating components (MOVs, Transzorbs, 
etc.) in favor of  an energy storage scheme followed by a slower 
dissipation process.  The initial transient charge and its 
associated energy are stored in capacitors, and eventually bled off 
to the neutral conductor through dissipative resistors.  This avoids 
the stressful instantaneous heating of the dissipative devices which 
can damage them, even to the point of explosion or fire.  Instead, 
the dissipation occurs over a longer period and therefore in a much 
less violent manner. 
 
The strategy of protecting the power line back to its neutral rather 
than to the safety ground prevents charge transients from flowing in 
the safety ground which also happens to be connected to the signal 
ground for connections to printers, modems, etc.  This avoids 
developing large transient ground voltage differences between these 
units, which helps avoid damage of their communications circuitry.  
The down side is that there is no protection for the so- called 
longitudinal mode transient which enters equally on the hot and 
neutral wires, and cannot be shunted to anything other than safety 
ground.  But in any case, for point-of-use surge protectors such 
protection is problematical because the behavior of the safety 
grounds is not  well controlled, as it can be for entire site or 
building protection. 
 
In summary, I believe the Zero Surge ZS 1800 to be a well engineered 
product, and the sales claims not to be overstated, as is so often 
the case.  It appears to work well in practice, and is of excellent 
quality.  Finally, its price is very reasonable.  I would call it a 
very good investment. 


