[sdiy] power line transient protection

[J. Larry Hendry]

I have worked in the power industry for many years and can tell you that I
do not believe there is any 100% effective electronic solution to lightning
protection.  However, solid state phase to ground arrestors are the most
commonly applied on the high voltage system.

I can tell you the most common cause I see for lightning damage in homes is
poor grounding practices.  I have seen several examples like the one below
where a strike some distance (never that far though) away  has caused
damage.  While it is possible for the transient to travel along the power
line and into the home, it is unlikely.  The arrestors on the power line, at
the step down transformer, and the step down transformer it self all serve
to mitigate these.

While I am no lightning expect, I can offer this simple explanation.  During
a lightning strike the current must obviously dissipate through the ground.
The ground, not being a perfect conductor offers resistance to this current
flow.  Therefore, voltage drop accumulates across the ground.  The voltage
at the strike point is raised above normal ground voltage and it migrates
back to normal the further you get away from the strike.  The difference
between 2 voltages at difference points on the ground is called step
potential in the power industry.  It can be thousands of volts per foot
during a lightning strike.  Livestock can be killed by current flow induced
through their bodies because of the difference in potential between their
front and rear feet.

Anyhow, back to houses.  Many times, I have found that houses have multiple
earth grounds.  This is not a good practice unless all the grounds are
solidly bonded together with wire of significant size.  There should be only
one ground, and everything should be grounded to that in a star
configuration.  In the USA, it was common practice of the phone company at
one time to ground their equipment on an incoming metal water pipe if that
was closer than the power ground.  Cable TV did whatever they pleased.
People installing home antennas, often installed arrestors on the signal
line and put a ground rod at the base of the antenna.  The problem is that
all of these ground attachments have the potential to be thousands of volts
different than each other during a reasonable close lightning strike.  The
problem in the house them becomes obvious as we start connecting things
together.

So, I offer 2 pieces of advice for electronic equipment protection:

1. check your grounding.  Have ONLY one ground rod, or bond them all
together with # 2 or larger wire.  Of course, check local electrical code.

2. Install a disconnect switch for your electronic gear.  I have a special
circuit run to me gear from my mains that goes through a disconnect switch.
It is only in the ON position when I am using it.  So, during storms, it is
OFF, ad therefore, no transient damage can occur.  Of course, this will not
protect the person who uses his gear during a storm.  :)  Essentially, this
is no more than unplugging stuff.  But, much handier and therefore more
likely to be actually done.

Larry Hendry

----- Original Message -----
From: Martin Czech <czech at Micronas.Com>
To: <>
Sent: Tuesday, April 10, 2001 7:48 AM
Subject: [sdiy] power line transient protection


Last year a friend lost his VCR, computer and TV due
to lightning. It wasn't even a direct hit, maybe a km
away. This is a frightening idea for us, because we
can't just pop into the next shop and replace our
destroyed diy synths.

Now this friend uses a line protection plug.
A quite expensive one and now I opened it and tryed
some reverse engineering...

Components:
standard rf line filter
two neon bulbs
two varistors
one gas filled discharge tube
two ptc or ntc elements glued to the varistors
a fuse

I guess EUR 10 worth, the price was 45 EUR!

This is what I think now:
The varistors block neutral and phase to protective earth.
The gas tube blocks neutral and phase.
One neon lamp simply states, that the device is working
(has power on outlet). But the other seems to indicate
some fault, I think cracked varistors due to huge transient.
I guess such a varistor will get hot and therefore
turn on the ntc, and the red light neon will turn on.

(... oh Roxanne... turn on the red light for me ....)

I guess this is industry standard. Others recommend
a high power zener diode instead of varistors.

Do you have any other ideas?

m.c.

p.s.: this message was composed with a new mail tool.
Problems? Give me a note.