Modem lightning protection

If someone really tries to use an acoustic coupler these days, make sure it works with at least 5-10 mm spark gap distances to be of any use.

The phone circuit is capable of delivering about 60 V very small current (on hook) and 20-60 mA (off hook) with a few volts, so you might be able to power the line side from the phone exchange and use fibers for data communication.

With a AC/DC tube TV, the chassis was directly tied to on of the mains connectors and depending on the polarity of the mains plug, directly connected to Neutral and hence Ground. A lightning surge might hence evporated the tuner, but then got an extremely low impedance path to mains ground, bypassing any other electronics.

During strong snow or hail storms, quite a lot of electric charge will build up in antennas.

I once observed that a dipole made for 80 m communication with the antenna coaxial disconnected from the transceiver but laying on the floor exhibited frequent (every few seconds) flashovers across the cable "UHF" connector between center tab and shield.

So if you want to avoid problems with electrostatic voltage (and noise) buildup in wire antennas, you really need to provide some DC current discharge paths to void such kV static charge DC build up.

Take a look how the "big boys" are doing it.

Assume that you want to measure the current in a 100-800 kV HT mains line. The actual measurement is done with a current transformer, but there is the common mode voltage problem (a few hundred kV :-).

One way of solving this is to use two fibers, use about 1 W optical laser power along one fiber to shine on a photo cell, which is powering the electronics sitting on the high common mode voltage and the other fiber is used to transmit down the actual measurement.

No,

The problems is that everyone thinks _only_ about their own grounding electrodes, but no-one thinks about the whole system grounding

This is what I tried to say about equipotential binding.

Have you looked at some of these destroyed modems etc. There are quite clear flashover paths between the telephone world and the mains worlds.

A grounding electrode with say 1 ohm grounding resistance is more or less useless for absolute potential nullification, but some equipotential bonding with quite mediocre ground electrode will save a lot of lives.

With a 10 kA lightning pulse the house potential will rise to 10 kV, but what is the problem, if everyone moves to that potential. Only external devices, such as modems will suffer.

That is the point of equipotential bonding,

I do apologize, Matt

I do apologize, Matt. I don't know why a kept reading "disagree". I read it several times (including twice just now) and still came up with "disagree".

Yes.

Huh? The fact is that this is a problem that *has* been beaten into submission. ...a *long* time ago. If he's still having trouble, he has a problem. It can be fixed.

It's not what you're saying.

I never said it can't happen. I'm saying that it doesn't (often) happen if everything is according to spec. The problem has largely been solved. It's not magic.

If you have two grounds in the house, you're inviting a nearby strike to travel *though* your house - not a good situation. Cork on the ocean.

Nothing will protect against a 10kA direct hit. If the is strike on the CO (your example), it won't make it to your house with anything like your numbers. It won't damage your modem if the CO demark is grounded to the same as the power. Cork on the ocean.

Then WTF are you whining about? Even that isn't the end-all. You don't want more than one ground (see above).

That was nice! Thanks Dimitrij.

George H.

Interesting, The previous owner of my house put a ground rod in at the garage... the opposite side of the house from AC line and phone entry. (he had a CB/police radio scanner and antenna.) The rod is still there but I cut the ground wire... maybe that was a good move?

George H

Yup, you really don't want earth currents from nearby lightning strikes passing THROUGH your house. The currents can be tremendous.

It probably made good sense when he had an antenna at that location.

Jon

This is totally excessive. I have one gas tube arrestor near the phone entry point, and it has saved all the gear from several major strikes or induced currents. The phone company has to replace the guts of their demarc box every couple years and also the buried cable every decade or so due to arcing and punch-through of the insulation. I've never had to replace anything protected by the arrester. I have had to replace the arrester tube once.

Jon

Around here, two grounding rods is strongly recommended. I added another one because there was significant AC current flowing through the wire mesh in the stucco exterior.

Of course you want the ground wires to be sized correctly so they don't burn up when they're doing their job.

On a sunny day (Thu, 03 Jul 2014 00:03:08 +0300) it happened snipped-for-privacy@downunder.com wrote in :

Yep, was some sort of joke anyways :-)

Right, I powered the whole thing from the phone line, you need 2 optos, one with a darlington photo transistor (for Tx).

The old phone lines were designed for a carbon mike in series, basically a variable resistor.. that can be done by the darlington. You need some way to prevent you hearing yourself... this is usually done with a transformer. To listen you can just have the other opto LED in series and limit the LED current a bit. We have 48 V system here. Beware of the ring signal! That is a lot of AC on the line all of the sudden!!.

Here in the early tube color age, the sets used a bridge rectifier, so it chassis was _always_ live, no matter what side was neutral. I had one of the first video recorders (a Philips LDL1000), and to bridge the mains gap so to speak, they modulated at 38 MHz IF and you had to modify your set to insert that after the tuner, modulator box came with that recorder.

One day (I lived on the first floor) there was a thunderstorm (not a big one). I had the window open and like to look at those things. The day before I had installed a new FM dipole on the roof for my neighbor...

Then I heard a bang, and a big white ball, about 30cm in diameter, glowing white, slowly descended in front of my window, stopped, hanged there, we looked at each other for many seconds, no heat, no radiation felt, just white, no structure, then it slowly descended down where it exploded with a load bang on the antenna of my downstairs neighbor who had the wire in between 2 trees. Next morning we went there, and only the ends of the wire were left, the rest was gone. Lucky he had unplugged it from his radio.

Ball lightning, seems very few people have seen that, always kept with me, .electrons... electron black hole?? Murat Ozer?

In Finland it is required (for new buildings) that you connect an uninsulated copper wire to your main ground bar, bury the cable all around the perimeter of the building and the other end is returned back to the same ground bar. This has the advantage of keeping any lightnings hitting a nearby tree from going through the roots into the basement and into some metallic tubes or wires inside the house.

This conductor is supposed to be a continuous run (no joints allowed) and even if the conductor might be broken sometimes later, there is still quite low grounding resistance due to the two connection to the grounding bar.

A simple ground rod is quite lousy in dry sandy environment, having quite high grounding resistance.

At least in Europe combined mains and phone surge protectors are sold. Of course, such devices needs to be connected to a grounded mains socket to work properly. Plug the device into a grounded socket, connect an extension cord with sockets for _all_ equipment such as computers, modems, printers, stereo etc. to this extension cord, route the phone line through this protection device to the modem. Some devices might also have surge protection for the antenna connection.

The equipotential point is now the PE connection in the wall socket.

Of course, it would be better, if the equipotential point would at the house entry of mains, phone, antenna, thus you wouldn't be so cautious about connecting audio etc. signal cables between different rooms.

Some UPS also have protected feed-through connections for phone lines.

That's probably a good, if expensive, idea.

How do you get a loop with no joints? ;-)

Here, there are requirements for the rods. They aren't (supposed to be) driven willy-nilly. Often, a few are driven (in the same area) to get the required ground resistance.

Drill a hole in a sphere or flat sheet.

Excellent information on surges and surge protection is at:

- "How to protect your house and its contents from lightning: IEEE guide for surge protection of equipment connected to AC power and communication circuits" published by the IEEE And also:

- "NIST recommended practice guide: Surges Happen!: how to protect the appliances in your home" published by the US National Institute of Standards and Technology

The NIST surge guide is aimed at the unwashed masses. The IEEE guide is aimed at people with technical background, which should be everyone here.

The NIST surge suggests the most equipment damage is from high voltage between power and signal (phone, cable,...) wires.

The basic protection points in the IEEE guide are

-- earthing system, which combines all earthing electrodes to a single system (isolated rods are an invitation to damage)

-- entry protectors for phone, cable, and other systems (like antennas and dishes). These entry protectors must be connected to a common bonding point on the earthing system with relatively short wires. An example of a wire that is far too long is in the IEEE guide starting page 30

-- a surge protector at the power service entry. Distance from the protector ground along the earthing wire to the common bonding point must also be relatively short.

-- plug-in protectors for sensitive electronics. All interconnected equipment must be connected to the same protector, and all wires (power, phone, cable,...) must pass through the protector. These protectors primarily protect by limiting the voltage from each wire to the ground at the protector. The connected equipment is on an "island" or "cork".

Along with someone else, I wonder if the phone entry protector is still working. I believe in the US these protectors are now gas discharge tubes, which should be pretty trouble free. (But GDTs have a short lag to protection.)

The entry protector used for cable in the US is usually just a ground block that allows the shield to be earthed. It does not limit the voltage on the center conductor.

The author of the NIST surge guide looked at the surge current that could come in on residential power wires. The maximum with any reasonable probability of occurring was 10,000A per wire. That is based on a 100,000A lighting strike to a utility pole adjacent to the house in typical urban overhead distribution. Service panel protectors with ratings far larger are readily available. High ratings mean the protector will have a long life. Recommendations for ratings are in the IEEE guide page 18.

With a large surge current to earth, the building "ground" potential can rise thousands of volts above "absolute" earth potential. Much of the protection is that all wiring rises together - the "cork" or "island".

If there is no service power panel surge protector, in US systems there is arc-over from the service panel busbars to the enclosure at about

6,000V. After the arc stabilizes the voltage is hundreds of volts. Since the enclosure is connected to the earthing system that dumps most of the surge energy to earth. The neutral is also connected to the earthing system which further limits the voltages in US systems.

The author of the NIST surge guide also looked at how much energy might be absorbed in a MOV in a plug-in protector. Branch circuits were 10M and longer, and the surge on incoming power wires was up to 10,000A (as above) - no service panel protector. The maximum energy at the MOV was a surprisingly small 35 joules. In 13 of 15 cases it was 1 joule or less. Protectors with ratings far higher are readily available. High ratings mean a long life. (One reason the energy was so low was arc-over at the service panel.)

Fuses don't work for protection because surge duration is far too short (and the fuse would be clearing against high voltage).

Thanks, The IEEE one is nice. I couldn't get the Nist link to work. I did find this page.

Which lists the "surges happen" handbook..3/4ths down the page but I couldn't figure out how to download it.

George H.

Not so expensive, if you install the loop before building the house:-)

The point is to avoid any outdoor or underground joints with dissimilar metals and in a few years, the electrochemical reactions will kill that joint. The main grounding bar is usually in dry conditions and with suitable mounting hardware the two ends of the loop can be connected to the main grounding bar without risk for electrochemical issues.