Lightning and Antennas

[Sparky67]

I live in a house on a hill, with antennas on the roof. 20 years ago we were struck by lightning - a side fork I believe - which destroyed most of the electrical equipment in the house including my amateur radio station. The smell of burning at 2am and finding none of the lights work, with two small children in the house, isn't something I would recommend.

The question is - do I disconnect all antennas when there is lightning about or do I fit an earthing system to the antennas and leave them connected? At the moment I disconnect the antennas and unplug everything from the mains to try to prevent a leader developing. Your thoughts?

[kellys_eye]

Usual practise was to leave the aerial open circuit and earth the receiver input. The idea being that lightning finds the shortest route to earth and if the aerial is 'floating' there's less chance of it attraccting a strike.

I suspect that nothing will prevent a direct lightning strike causing major damage though.

[Radio Wrangler]

One school of thought is that you have the antennae earthed, but do it outside the house, to a proper earth rod, then if the worst happens, there is a degree of isolation.

The idea is that spiky things are supposed to spray appropriate charges, discharging the air around them. Lightning conductors are really lightning preventers and have a sort of conical volume underneath them which is considered to be protected.

David

[G6Tanuki]

I recall seeing in some photos of pre-WWII amateur stations a big knife-switch used to earth the aerial when it wasn't in use.

You can buy 'in-line' coaxial lightning-protectors with SO239 sockets and a nut/bolt for attaching an earth-wire. To use one of these with a traditional wire antenna - remember that 4mm banana-plugs fit SO239 sockets!

http://www.hamradio.com/detail.cfm?pid=H0-006891 [MFJ stuff is available from plenty of UK ham-radio resellers too].

When I was running HF backbone links the standard inline thingy was a Polyphaser: http://www.dxengineering.com/search/...ing-protectors

Though if you get a direct hit then nothing much will save you [your flexible coax cable can go solid when the individual strands of the braid weld to each other...]

[Refugee]

I was told that when my father was very young their house got struck and the lightning just jumped from the end of the aerial wire to the radio and did the damage anyway.
Outside it blew the chimney stack off sending it into an upstairs room in the house next door.

[Billy T]

.
Most lightning damage is caused by induced voltages and currents, so disconnection/grounding etc can be an effective precaution to minimise such collateral damage in some cases.

However, the discharge current of a true 'strike' knows no barriers and will wreak havoc regardless of any and all 'domestic grade' precautions, and most commercial protections as well. In fact, it is arguable that lightning protection apropo domestic installations is a waste of time and it is better to simply disconnect any sensitive equipment and wait for the storm to pass.

Look up 'lightning strike damage' in 'Google Images' and you will see just how impotent any precautions taken may be, should nature decide to bite!

A few years back I was working overseas doing risk assessment on government buildings of a Commonwealth country and in the multistorey building housing their computer systems I saw that their lightning conductor was run from roof to basement inside a steel waterpipe. Following that pipe up to the roof, I then found that the lightning rod itself had suffered physical damage at some time and had been removed, so the connection had bee attached to a metal ventilator shaft. The site was in a known electrical storm area, so I advised that they begin an urgent review of the entire system and get started on a remedial program.

A few months of inaction later, they were hit by a major electrical storm and immense damage was done to their electrical and computer networks*. All that the pipe did was channel the strike energy down several floors to the server room, blowing out electrical/electonic systems on every floor as it went.


'Switch off, unplug and disconnect' is by far the best precaution you can take, and that should be done before the thunder rumbles, that is what weather forecasts are for. Bear in mind that a 'leader' usually goes up before the 'strike' comes down, hence 'hair standing on end' etc.

Billy

*As a final insult, the batteries for their extensive UPS systems were knackered as well, so their UPS had a run-time only slightly longer than the lightning strike.

[Radio Wrangler]

Lightning effects on ground installations have been studied, but not to the extent those on aircraft have.

Aircraft get struck by lightning relatively frequently. An arc attaches somewhere to the fuselage, current flows through the fuselage and an arc towards the ground starts from somewhere else. Usually pointy things are preferential attachment points - radio antennae!
Planes have static discharge wicks deigned to minimise the build-up of potential so landings aren't accompanied with huge zaps. The wicks are likely continuation arc points. Effectively the plane becomes a series connection along the path of a lightning strike.

A direct hit can melt antenna elements and cables. he hope is that the backup radio isn't damaged by the induced surge.

The strike passes a huge current through a metal plane. The evolution towards carbon fibre aeroplanes has a nasty consequence that the planes are becoming much higher resistance and the amount of energy dissipated along the current flow route increases massively., threatening major structural damage. Airframe manufacturers are having to incorporate metal mesh layers to restore protection.

There's been a lot of work done on lightning effects and there is a standards document setting out what performance a good radio (and other systems) installation must have and how to test it.

DO-160G is the latest revision. Chapter 22 covers the effects of direct lightning hits. Chapter 23 covers induced transients applied to radios when other parts are struck.

The direct stuff doesn't help us much down on the ground except for people living in fully enclosing all-steel houses (There are some of those in Kirkliston, built from recycled ship's plates!) But the indirect induced effects are a good guide.

Differential potentials on a struck airframe are expected to couple up to 125v/125A surges with a 40us rise time (0 to 100%) and then to fall with an 80us (100 to 50%) exponential decay. This can be positive or negative going. Apply this to every pin of every connector in both polarities.

Induction from the current path/wiring is expected to apply a damped sine ring to each connection to a radio. 250v/10A peak, both starting polarities. 1MHz for longer wires in large planes, 10MHz for small planes to represent typical resonant frequencies.

How do you protect against these? Well you certainly need to make sure no connection presents an impedance close to 1 Ohm, that'll match it to the 125V/125A transient, and it'll hit a peak power of 3.9 kilowatts! Ouch! The ring waveform only peaks at 600+ watts.

There are whacking great (by modern standards) Zener diodes called TVS diodes. The big 'SMC' sized ones will handle 1200W peak and routinely survive the tests when placed right at the input of a connection. If your connection can tolerate several tens of Ohms series resistor right at the connector, you can use smaller SMB or even SMA (no, not the connector) sized diodes.

The Apple MAC I'm typing this on had a blown-up ethernet port. Apple say it's £600 for a new motherboard. It looks like an antenna flashover dnced around the shack while I was away last year. An ethernet to 'thunderbolt port' adapter is an amusingly ironic cure which cost an awful lot less.

Grounding things in domestic radio installations runs into a whole lot of regulatory pain over PME earthing schemes.

David

[G4_Pete]

Whilst it may detract from the hobby and the feel of being able to adjust tuning ect the only low risk way if you are in a high probability lightening site is to put all the radio gear in a remote shed and fibre optic back to your house. In industry this is how we brought critical computer lines back in to the control houses. Everything else cabled came into the building via a huge earthed plate on the outside wall with appropriate earthing and lightening traps depending on the cable service. Secondary transient protection or earthing was then applied on signal lines as apropriate.

Unplugging cables on a storm approach is not without risk as potentials build up in the air prior to a full blown storm and you do not want to be holding the cable in one hand in this situation. I have had sparks jumping between the centre and braid on a sloping dipole with what I thought was a distant storm aproach.
Pete

[lesmw0sec]

It is always worth unplugging aerials when not in use. Here on Anglesey we often get a sudden increase in static with no warning and in the absence of a storm. I had to replace the bulb 'fuse' in an FT101 as a result of leaving a 160M dipole plugged in. Fortunately the fuse did its job well.

[G6Tanuki]

Originally Posted by G4_Pete

Everything else cabled came into the building via a huge earthed plate on the outside wall with appropriate earthing and lightening traps depending on the cable service.

That's the technique known as "ground windowing" - see para.8 of http://lightning-protection-institut...ng-protect.htm

[paulsherwin]

Before the 1960s most UK households had big professionally installed longwire aerials stretching across the garden or back yard. These were usually terminated by a substantial knife switch on the windowsill where the aerial entered the house, allowing easy isolation once the thunder started or when the radio wasn't in use.

[emeritus]

Interestingly enough, the 1936-37 Bulgin catalogue does not offer a substantial knife switch, but a compact "QMB Lightning Switch and Arrester", the effectiveness of which they were sufficiently confident to offer a "£100 guarantee of efficiency". I wonder if anyone ever made a claim? PDFs of the catalogue extract and the patent covering the switch attached.

[russell_w_b]

I guess if your AMU is configured with an inductor to earth at the aerial side or shunt loading coil arrangement you'd have a static leak to prevent a charge building up (but not a direct strike). Or perhaps you could connect a hi-Z choke to the aerial as a permanent static leak?

[Sparky67]

Thanks for the advice chaps. In a previous life I installed antenna systems on towers and have removed large antennas which have been struck by lightning several times - the damage to the finials was obvious - although no damage inside the building occurred (except one site where the BT line terminations got fried...!). On these sites the feeder braids were earthed at the top and bottom of the tower and just before the building ingress. The radio eqpt chassis were connected to their racks, the racks were bonded back to the tower (via 25x3mm tape) and the tower legs were cross-bonded and the whole system had a <15 Ohms resistance to earth. I was told the cross-bonding kept everything 'earthed' at the same potential during a strike and prevented damage.

I suspect to fully protect the radio station in the house would require a lot of earth tape, earth rods and cross-bonding. Think disconnecting and crossing fingers is going to be the way forwards for me...

[G6Tanuki]

You were lucky: one lightning-strike I had to deal with welded the sections of a 60-foot trailer-mount telescopic "Versatower" together. The strike also catastrophically melted the winch-cable. When I got to the site the mast was half-telescoped-down and firmly stuck like that, going nowhere.

The wheel-bearings on the trailer were also welded solid! We cut the lot up on-site using angle-grinders and oxy-acetylene, cashed it in as scrap, and claimed on the insurance.

[peter.sables]

In the Army serving in Malaya. We had lots and lots of ground strikes. We saved the coax and set by this little tabaco tin. Simple but it worked (most of the time.

Pete G4MRU

[QWENIX]

Never ever connect the antenna to earth. Leave the circuit open. It's wise to extra discharge the antenna after the lightning weather has gone. So no built up voltage charge could damage the equipment by discharging through the equipment to earth. Only the mast could use a lightning conductor.

qwenix

[peter.sables]

Hi.
I think maybe you have missed the point. The shellac is covering the bearing So the antenna is not earthed. The shellac is regarded as one micron thick at the top of the bearing. So the antenna is not earthed. But the smallest heat will short out the antenna very quickly. Maybe a little "heath Robinson". The bottom of the bearing has no shellac on it. ie earthed. We also had a flat carbon block, once again with shellac covering one face of it.

Pete

[Peter.N.]

I'm afraid I do nothing with my ham antennas, I don't even have an earth in the shack as my only HF aerial at present is a 200' balanced fed dipole which I can tune up on any band, my 2m ant is at about 40' and we are 500'ASL.

I was looking out of the window during one storm and the 11kv power line was struck about 2 poles down the hill from us which must be at least 50' lower. I did have my router blown up by a strike but that was further down the road.

Peter