DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=googlemail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type;
        b=uGHcyOyrBirrfqd2XRN/YE/yBUPhYPAz+8Gd41RB/ggNp1rSnSzcoGAqADqsMnoi1B
         XO4FL8z7qFFr1+U//WyUAjpNZ7fPyyp0LsLErmQFWTyLcHmeRvudjAG+nKqWEb+1ucyQ
         x6insZPoCjazIAWBawWoodNd91oQ+IoD4Mutw=
MIME-Version: 1.0
In-Reply-To: <op.vjhyi5kpyzqh0k@pc-roelof>
References: <000c01cb5a7a$44cfcc50$0401a8c0@xphd97xgq27nyf>
	<4C1A271605FB0B62@smtp208.alice.it>
	<op.vjf78o2pyzqh0k@pc-roelof>
	<4C9A8CC5.30305@gmx.com>
	<e883a371b121f7efe4940eeacc98be35@xs4all.nl>
	<4C9B3FCE.5080700@gmx.com>
	<op.vjhyi5kpyzqh0k@pc-roelof>
Date: Thu, 23 Sep 2010 20:27:35 +0100
Message-ID: <AANLkTimay6khcCa2R__C5O3qucRQA6UCFy28Htv38wfC@mail.gmail.com>
From: Andy Talbot <andy.g4jnt@googlemail.com>
To: rsgb_lf_group@blacksheep.org
DomainKey-Status: good (testing) 
Subject: Re: LF: PA0RDT miniwhip in TX?
Content-Type: multipart/alternative; boundary=0016361e883221a28a0490f2415c
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org


--0016361e883221a28a0490f2415c
Content-Type: text/plain; charset=ISO-8859-1

Lightning strikes to ground are, I believe, by far the most damaging type.
Many kV / metre, even approaching MV/m can be induced in teh ground.
People standing with feet apart in line with the strike get electrocuted
from current flowing up one leg and back down the other to a point only
400mm apart.  I know someone who saw this actually happen on a football
pitch!

This transient (of a few microseconds max) voltage gradient induces massive
currents that get coupled into buried and overground cables, travel along
them as a transmission line over (under) ground until they reach the
target.   The sharp edge impulse of several kV hits the equipment wrt. to
chassis, travels across transformer capacitiance or breaks down filter
components that are supposed to decouple to ground, and sparks across PCB
gaps.    The regulatory 5mm (I think) gap between mains line and equipment
side is for this very reason - but may not always be adequate.

On our microwave beacons   www.scrbg.org located on a remote hilltop site we
took great pains to ensure every single bit of metalwork was bonded very
firmly together, this included antenna, mast, equipment rack and the mains
input cable armouring.   Then a ring of earth stakes was put in at a few
metres distance all round the cabin.   These were bonded with thick bare
buried copper wire run back to the cabin and connected to the metalwork at a
common point.   The aim was to create an equipotential zone in the ground
around the cabin so lightning induced transients would pass around it.
It all worked very well for several years worth of thunderstorms,  Until one
day the beacons all suddenly went off after a local storm, and did not
return when mains power was restored locally.   Going off with power outages
was reasonably common, but they always came back

On investigation, the switchmode master PSU/Battery charger had failed,
A filter capacitor from live to ground had blown apart and a diode in the
input bridge had blown shorted   What we think happened is that there was a
ground strike somewhere near the 1km run of undergound cable.    This
induced massive currents in the cable armouring, which because it was high
resistance steel, caused the EM field to penetrate into the L/N mains
conductors. (It may also not be properly grounded at the supply end)
Because the mains is inherently unbalanced, (at the 11kV/415V transformer on
the supply pole the neutral must be grounded somewhere)  the common mode
transient got turned into a differential mode one that travelled back up the
cable to pop the diode.   The capacitor probably went due to the common mode
component.

Furtehrmore, the battery that would keep the beacons running for several
hours ended up so flat it was permantly damaged and useless.

After replacing the PSU, a new mains input filter box was added at the point
the underground cable enters the cabin.   This is built into in a dedicated
diecast box bonded firmly to the earthing bus,  with two  large transorbs
(varistors) from line and neutral to case/ground, a filter block, and
another pair of transorbs.   More transorbs go between L and N.   Transorbs
are designed to absorb huge energy spikes from overvoltage surges - the very
thing.
In addition, a hardware 12V watchdog was installed, to permanently switch
everything off when battery voltage falls below 10V to protect the battery
itself.   This operates over and above the software monitor that just
switches the beacons into a lower power standby mode when mains failure is
detected.  It was the extended period of standby that killed the battery

We've had no more problems  - but there haven't been so many thunderstorms
as there were a few years ago.  There have been a few outages, but full
operation always returned.  The hardware 12V watchdog was triggered just
once by a longer power outage.   The advantage of a decent telemetry system!

Andy
www.g4jnt.com


On 23 September 2010 18:50, Roelof Bakker <roelof@ndb.demon.nl> wrote:

> Hello Mike,
>
> My antenna was mounted 15' off the ground with a ground wire connected to
>> the coax braid right under the antenna.
>>
>
> That is interesting. This might imply that the ground at the antenna is at
> the voltage level induced in the 15' long wire to ground. This might lead to
> a number of unforeseen effects. So, I wonder if the FET was damaged due to
> induced voltage on the PCB antenna or due to the voltage over the wire to
> ground.
>
> I have mounted the antenna at about the same height, but the shield of the
> feedline is connected to a ground stake at the bottom of the mast. I use a
> BNC panel mount feedthrough like this one:
>
> http://rocky.digikey.com/weblib/Emerson/Johnson/Web%20Photos/CP-AD-556M.jpg
>
> It is mounted on a small piece of aluminium sheet, which is connected with
> a short wire to the ground stake.
> The feedline is broken up at the bottom of the mast and two BNC connectors
> have been fitted.
>
> Though real heavy thunderstorms are rare in this part of the country, I
> have had strikes within 300 m with no ill effects on the antenna.
>
> 73,
> Roelof, pa0rdt
>
>

--0016361e883221a28a0490f2415c
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div>Lightning strikes to ground are, I believe, by far the most damaging=
 type.=A0=A0 Many kV / metre, even approaching MV/m can be induced in teh=
 ground.=A0=A0 People standing with feet apart in=A0line with the strike=
 get electrocuted from current flowing up one leg and back down the other=
 to a point only 400mm apart.=A0 I know someone who saw this actually happ=
en on a football pitch!</div>

<div>=A0</div>
<div>This transient (of a few microseconds max) voltage gradient induces=
 massive currents that get coupled into buried and overground cables, trav=
el along them as a transmission line over (under) ground until they reach=
 the target.=A0=A0=A0The sharp edge impulse of several kV hits the equipme=
nt wrt. to chassis, travels across transformer capacitiance or breaks down=
 filter components that are supposed to decouple to ground, and sparks acr=
oss PCB gaps.=A0=A0=A0 The regulatory 5mm (I think) gap between mains line=
 and equipment side is for this=A0very reason - but may not always be adeq=
uate.</div>

<div>=A0</div>
<div>On our microwave beacons=A0=A0 <a href=3D"http://www.scrbg.org/">www.=
scrbg.org</a> located on a remote hilltop site we took great pains to ensu=
re every single bit of metalwork was bonded very firmly together, this inc=
luded antenna, mast, equipment rack and the=A0mains input cable armouring.=
=A0=A0 Then a ring of earth stakes was put in at a few metres distance all=
 round the cabin.=A0=A0 These were bonded with thick bare buried copper=A0=
wire run back to the cabin and connected to the metalwork at a common poin=
t.=A0=A0 The aim was to create an equipotential zone in the ground around=
 the cabin so lightning induced transients would pass around it.=A0=A0=A0=
=A0 It all worked very well for several years worth of thunderstorms,=A0=
 Until one day the beacons all suddenly went off after a local storm, and=
 did not return when mains power was restored locally.=A0=A0 Going off wit=
h power outages was reasonably common, but they always=A0came back </div>

<div>=A0</div>
<div>On investigation, the switchmode master=A0PSU/Battery charger=A0had=
 failed,=A0=A0=A0=A0 A filter capacitor from live to ground had=A0blown ap=
art and a diode in the input bridge had blown shorted=A0=A0 What we think=
 happened is that there was=A0a ground strike somewhere near the 1km run=
 of undergound cable.=A0=A0=A0 This induced massive currents in the cable=
 armouring, which because it was high resistance steel, caused the EM fiel=
d to penetrate into the L/N mains conductors.=A0(It may also not be proper=
ly grounded at the supply end)=A0 Because the mains is inherently unbalanc=
ed, (at the 11kV/415V transformer on the supply pole the neutral must be=
 grounded=A0somewhere)=A0 the common mode transient got turned into a diff=
erential mode one that travelled back up the cable to pop the diode.=A0=A0=
 The capacitor probably went due to the common mode component.</div>

<div>=A0</div>
<div>Furtehrmore, the battery that would keep the beacons running for seve=
ral hours ended up so flat it was permantly damaged and useless.</div>
<div>=A0</div>
<div>After replacing the PSU, a new mains input filter box was added at th=
e point the underground cable enters the cabin.=A0=A0 This is=A0built into=
 in a dedicated diecast box=A0bonded firmly to the earthing bus, =A0with=
 two=A0 large transorbs (varistors) from line and neutral to case/ground,=
 a filter block, and another pair of transorbs.=A0=A0 More transorbs=A0go=
 between L and N.=A0=A0 Transorbs are designed to absorb huge energy spike=
s from overvoltage surges - the very thing.</div>

<div>In addition, a hardware 12V watchdog was installed, to permanently sw=
itch everything off when battery voltage falls below 10V to protect the ba=
ttery itself.=A0=A0 This operates over and above the software monitor that=
 just switches the beacons into a lower power standby mode when mains fail=
ure is detected.=A0 It was the extended period of standby that killed the=
 battery</div>

<div>=A0</div>
<div>We&#39;ve had no more problems=A0 - but there haven&#39;t been so man=
y thunderstorms as there were a few years ago.=A0 There have been a few ou=
tages, but full operation always returned.=A0 The hardware 12V watchdog=A0=
was triggered just once by a longer power outage.=A0=A0 The advantage of=
 a decent telemetry system!</div>

<div><br clear=3D"all">Andy<br><a href=3D"http://www.g4jnt.com">www.g4jnt.=
com</a><br><br><br></div>
<div class=3D"gmail_quote">On 23 September 2010 18:50, Roelof Bakker <span=
 dir=3D"ltr">&lt;<a href=3D"mailto:roelof@ndb.demon.nl">roelof@ndb.demon.n=
l</a>&gt;</span> wrote:<br>
<blockquote style=3D"BORDER-LEFT: #ccc 1px solid; MARGIN: 0px 0px 0px 0.8e=
x; PADDING-LEFT: 1ex" class=3D"gmail_quote">Hello Mike,<br><br>
<blockquote style=3D"BORDER-LEFT: #ccc 1px solid; MARGIN: 0px 0px 0px 0.8e=
x; PADDING-LEFT: 1ex" class=3D"gmail_quote">My antenna was mounted 15&#39;=
 off the ground with a ground wire connected to the coax braid right under=
 the antenna.<br>
</blockquote><br>That is interesting. This might imply that the ground at=
 the antenna is at the voltage level induced in the 15&#39; long wire to=
 ground. This might lead to a number of unforeseen effects. So, I wonder=
 if the FET was damaged due to induced voltage on the PCB antenna or due=
 to the voltage over the wire to ground.<br>
<br>I have mounted the antenna at about the same height, but the shield of=
 the feedline is connected to a ground stake at the bottom of the mast. I=
 use a BNC panel mount feedthrough like this one:<br><br><a href=3D"http:/=
/rocky.digikey.com/weblib/Emerson/Johnson/Web%20Photos/CP-AD-556M.jpg" tar=
get=3D"_blank">http://rocky.digikey.com/weblib/Emerson/Johnson/Web%20Photo=
s/CP-AD-556M.jpg</a><br>
<br>It is mounted on a small piece of aluminium sheet, which is connected=
 with a short wire to the ground stake.<br>The feedline is broken up at th=
e bottom of the mast and two BNC connectors have been fitted.<br><br>Thoug=
h real heavy thunderstorms are rare in this part of the country, I have ha=
d strikes within 300 m with no ill effects on the antenna.<br>
<br>73,<br>Roelof, pa0rdt<br><br></blockquote></div><br>

--0016361e883221a28a0490f2415c--