Where damage is not acceptable, that separation between protector
and electronics increases protection. For example, you telco connects
their $multi-million computer directly to wires all over town. It may
suffer about 100 surges during every thunderstorm - without damage.
To be effective, a protector is located up to 50 meters from the
computer AND as close as possible to earth ground.

What provides protection to your furnace or dishwasher? Not the
protector. Protection is earth ground. The effective protector
connects surges to earth ... better when distant from the appliance.

Why is an H-N protector ineffective? Surge on both wires ignores an
H-N protector; continues on to earth ground destructively through a
furnace. Effective protection connects each wire (H N G) to earth
(not to each other). Separation between protector and furnace means
even less surge finds earth via the furnace. Others also describe how
and why a 'whole house' protector is so effective because it makes
that shortest connection to earth.

Where does surge energy get dissipated? Any protector that somehow
makes surge energy disappear (ie H-N) is mythical. From the NIST:
What these protective devices do is neither suppress
nor arrest a surge, but simply divert it to ground,
where it can do no harm.
That surge energy must be dissipated. Effective protection means
energy is dissipated harmlessly in earth.

Why does separation increase surge protection? Longer wire
increases impedance. Separation means a surge is less likely to find
earth ground via the furnace. Shorter wire (protector to earth) means
less impedance (as BillGill, et al have noted). Catastrophic surge
must obtain earth via a breaker box earth ground rod; not
destructively via the furnace. More reasons why a protector is only
as effective as its earth ground and why MOVs too close to the furnace
don't provide effective protection.

Appliances contain internal protection that make smaller surges
(temporary overvoltage) irrelevant. Why is one 'whole house'
protector so effective? A catastrophic surge earthed before entering
a building will not overwhelm protection inside an appliance or
furnace. To be effective on catastrophic surges, a protector needs
that short (ie less than 10 foot) connection to earth ground. A
protector is only as effective as its earth ground. Even better, the
superior protection costs about $1 per protected appliance; not a
ridiculous $25 or $150 for a plug-in protector.

On May 21, 7:02 pm, Nate Nagel wrote:
I'm ASSuming that I already know the answer to this, but I'll try anyway
- assume a usual littlesurgeprotectorwith three MOVs, H-N, H-G, N-G.
Will it provide protection to a device connected directly by hard
wires to the receptacle that it's plugged into? I'm guessing that it
really doesn't care as to "upstream" or "downstream" only that the level
of protection depends somewhat on the distance from theprotector. Also
if asurgedestroys it it obviously will not disconnect a device not
"downstream" of it.

Reason I ask is, due to the fun and games I've been having with
repairing appliances due to a bigsurgeabout a week and a half ago, I
thought that addingsurgeprotection to my furnace and air filter would
be a good idea. Problem is, now that I dig into it, the furnace is
hardwired to the back of a box on the side of the furnace. There is a
switch and a duplex receptacle in that box. The switch controls the
furnace and one half of the recep; the other is always hot. A
humidifier is plugged into the switched side, a condensate pump into the
unswitched side. I figured the best I could do, without adding some
kind of hardwiredsurgeprotection, was to plug it into the unswitched
recep for the condensate pump and it would provide the same protection
as if it were inline, with the caveat that if asurgedestroyed the MOVs
in thesurgeprotector, it is possible that it might zap the furnace
before the breaker tripped. Am I correct?

Are there any common, commercially available point of usesurge
protectors designed to be mounted in, say, a 1900 box screwed to the
side of a piece of equipment? I am thinking that one at the dishwasher
might be advisable as well, as its only protection appears to be one H-N
MOV and we already established that that wasn't sufficient in at least
one instance