[time-nuts] GPS antenna selection — lightning

Eric Scace eric at scace.org
Fri Aug 5 10:37:28 EDT 2016

Unfortunately, an antenna, cable, or piece of electronics located indoors is just as susceptible to lightning surges as one that is outdoors.

Lightning-induced surges couple into these systems electromagnetically across a wide range (VLF to SHF) of frequencies. When you think about your home from an electromagnetic viewpoint, just imagine your structure with all non-conductive materials absent. For a typical wood or brick/stone house in North America, what you are left with is:
metal plumbing pipes and fixtures, with their geometry suspended in space
house wiring, CATV, Ethernet, and telephone cabling, and their service drops, all suspended in space
electrical & electronic circuits of every kind (WiFi note, computer, appliances), their power supplies and AC power cords, also suspended in space
metal furniture? That’s hanging out there, suspended in space, too.
any I-beam or other steel structural elements, some random aluminum flashing, door knobs, and other similar metal construction materials used in the home.
That is what an electromagnetic pulse sees as it approaches and sweeps over your home… all hovering over a lossy ground plane (earth) its varying dielectric constant.. Each one of those pieces of metal, hanging in space, is an unintentional antenna that experiences voltage differentials and current flows.

A GPS antenna and its coax line that is installed next to a window is no different from the same antenna/coax installed one meter outside the window… or 10 meters away outside the window. All three installations are effectively “outdoors” from an electromagnetic viewpoint, and all three need effective surge protection from lightning-, cloud-, and precipitation-induced voltage surges.

(N.B.: Snow can be particularly bad for voltage surges. I’ve seen thousands of volts per meter potential differences in moderate-to-heavy snowfall that produced very significant current flows on cables.)

Surge protection for your antenna, its attachment to your receiver(s), AC/DC power supply lines, and any other signal lines of significant length is cheap insurance.

My continuously-operating electronics lives in an enclosed rack cabinet — not too much worse than a proper Faraday cage. Every cable entering the cabinet has surge protection at the point of entry. The cabinet is bonded to earth ground by 2” copper flashing. In the past this system lived 22 years on a mountaintop home, 1200 ft above surrounding terrain. Lots of thunderstorms — zero damage/disruptions during that time… a sample size of one, admittedly, but during the first 18 months at that site I had two lightning-surge damaging events before I got serious about protection.

I have equipment at a coastal site with multiple 130-ft towers. That site had damage events every 2 years or so — even when cables to the “outside” were disconnected, and AC mains power was shut off at the main circuit breaker box. After implementing comprehensive surge protection, we have had zero damage over the last 12 years.

— Eric

> On 2016 Aug 04, at 19:46 , Bob Camp <kb8tq at n1k.org> wrote:
>>> Grounding the antenna is always a good idea.
>>> A surge suppressor in the line could save you some
>>> real cost if there is a lightning strike.
>> I did a quick search for SMA/BNC/TNC based surge
>> protectors and not much did come up, any suggestions
>> what to use there?
> There are a *lot* of them on eBay. Many of them have N connectors on them.
>>> I don’t know about Austria, but here in the US,
>>> both are required.
>> Outside definitely, "inside" I'm not sure, but it
>> won't hurt to have additional protection for the
>> receiver(s).
> It is a good bet that the antenna will be outside. I’d plan it that way.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://www.febo.com/pipermail/time-nuts/attachments/20160805/3a47e4e0/attachment.sig>

More information about the time-nuts mailing list