On Fri, 2006-05-05 at 10:19 -0400, Randy K4QO wrote:
> With apologies to "Field of Dreams", what I mean is that just grounding
> things sometimes means that you are inviting lightning to go to ground
> there... through your equipment.
The rules on ground since at least 1916 depend only on grounds as if the
earth was a ten foot thick slab of silver. It isn't and its not all that
good a conductor though its better wet than dry.
>
> In my line of work with commercial radio towers, I have found that
> sometimes it is a mistake to ground certain items. Improvements in
> survivability were gained when certain protection points were removed.
> For example, a serial communication line was protected with a device and
> was always getting fried along with the device it was hooked to.
> Removing the protection actually caused the devices to survive for more
> than three years with no problems.
One scheme that seems to work better is to surround the equipment by a
very good conductor. A ribbon of copper 18" or a couple feet wide all
the way around the building, tied to all the grounds with power and
coaxes as well as phone lines passing through arrestors mounted IN that
ribbon.
>
> At my QTH, I look for ways to remove paths like was described by the
> "RIP" post. Details like replacing switcher wall warts with
> transformer based ones provide a slight measure of advantage. The
> transformer ones have (admittedly small) isolation from ground and the
> AC lines. This isolates the router and cable modem (which is attached
> to the cable!) from the AC mains and doesn't "invite" the surge to use
> this path to go to ground. I won't go into a full treatise about
> grounding but if you take the time to personify lightning surges and see
> what paths it can take to ground (via your AC power plugs), you might
> see some opportunities for isolation. One more thought - I use a large
> linear supply with better transformer isolation to power the cable modem
> and router and my radio supplies are all linears for the same reason.
> No switchers for me!
>
> BTW, I have heavy surge arrestors at the cable entry point with ferrites
> on the cable on the house side of things to encourage the surges (that
> are basically an RF signal) to go to ground at the points I have
> provided outside the house. Its my own brand of voodoo but it is based
> on observations garnered at the commercial sites.
Depending entirely on grounds for lightning protection is like filtering
with a single pole filter, e.g. low pass filtering with an integrator.
Yes, it will show a gently sloped cutoff. Filtering with just a
capacitor. That filter capacitor is much more effective if there's some
series impedance, even better if that series impedance is an inductor.
Likewise lightning protection depending entirely on grounds works after
a fashion, a poor fashion, adding some series impedance (inductor, or
air gap, or resistor) makes that ground more effective. Grounding the
tower separately, then adding impedance to the coaxes and using a
separate shack ground makes a PI attenuator that is more effective so
long as the two grounds don't get coupled through mutual sort of
conductive earth. And those series gaps are really effective when they
are a couple feet. I've lost stuff when the series gaps (e.g. coax
separated from patch panel) weren't large enough. What stops 120
VAC is no bother for a few KV to leap. or a few tens of KV.
>
> This is a topic fraught with lots of opinion and some science and I
> don't mean that you shouldn't ground your station. The commercial sites
> frequently have some of the best grounding known to man and still see
> damage. Of course, at the end of the day, your best defense is to
> totally isolate your prized possessions during the storms.
>
> 73,
> Randy
> K4QO
--
73, Jerry, K0CQ,
All content copyright Dr. Gerald N. Johnson, electrical engineer
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|