DRJ's Spring 2019

Conference & Exhibit

Attend The #1 BC/DR Event!

Winter Journal

Volume 31, Issue 4

Full Contents Now Available!

Tuesday, 30 October 2007 05:53

Getting Your Ground Down to Earth

Written by  Donald E. Kimberlin
Rate this item
(0 votes)

There it was...the “perfect” pay station location, at a gas station on the corner of a busy intersection, but off on the corner of the property so people could drive up to it and make their calls. It was so accessible and visible it should have attracted several times the average number of local calls. It should have returned its investment in one-third the typical time.

But it didn’t. In fact, it was out of order more often than it was in service. No physical damage occurred. No vandalism, no cars driving into the booth, no floods. It never had a real line outage. Its own advanced electronics never reported a failure to you. You never found out it failed until someone at the location called to complain.

The electronics kept dying. And every time you replaced them, the new ones went dead. No smoke, no flame, no flash. Just dead. Well, somebody said it might be lightning, so you bought surge protectors for the power line and phone line, but to no avail.

Somebody said you must have a bad ground, so you hired the oldest, most skilled electrician you knew, and he thoroughly inspected the power line ground and declared that it fully met the National Electrical Code. He even knew how to use a “megger” and confirmed the ground resistance was so low it was almost unreadable. The installation met every safety requirement in the book, he told you. But the phone still kept dying.

The words the electrician used contain the seed of what is wrong. By making the installation safe for people, he also created an enormous hazard for solid-state electronics. The reason: the ground those electronics need has to be intimately connected to the potential of the ground right where the phone booth is, neither back at the building’s power panel nor back at the telephone exchange.

You suggested driving a ground rod at the booth, and the electrician told you that you’d violate the code if you did that. Yes, if you did that and that alone, you would be violating the code and creating a hazard for people. The white or green wire ground used for power distribution and safety purposes, if broken and replaced by another path to ground (like your proposed ground rod), could let the power ground in your equipment rise 115 volts above the ground that people stand on in the booth and shock them just as if they stuck a hairpin in a light socket.

What the electrician didn’t remember from his exam days was the special conditions in the national Electrical Code for communications equipment. They take the form of footnotes that refer to the use of an “alternative earthing electrode.” That’s the fancy name for your ground rod at the isolated booth.

When you read about the alternative earthing electrode, you find that the code simply requires that you “bond” that rod to the power ground point back in the building with a No. 6 insulated “bonding conductor.” It is very large to ensure that it is physically durable and won’t break...or at least will break last. Its size also reduces its inductive reactance to make it rapidly follow changes in the ground potential.

“Changes in the ground potential”? But isn’t the ground always “at ground”? Isn’t it by definition a potential of 0 volts? Don’t we always rely on it to be an unchanging electrical platform? Well, yes, we do. But the truth is that when a huge amount of current passes through the earth right beneath us, the ground takes on a very jelly-like electrical consistency for those thousandths of a second while the current is passing by. For a time shorter than the blink of an eye, the earth we rely on to be so stable simply cannot stay at 0 volts. A pay station relying on a ground at the other end of a power line that is 50, 85, or 120 feet long to the power panel is connected to the end of an electrical “whipsaw” whenever a huge current passes through the earth.

When can this happen? Whenever a lightning bolt strikes anywhere within a mile or so of the pay station. The effect is just like a huge invisible wave radiating outward for up to a mile, until it dissipates. For that instant, everything in the vicinity is at a different ground potential than the power safety ground.

Visualize being in a small boat with a rope tied back to the shore. Suddenly, a load of rocks gets dumped in the pond and creates a 6-foot wave. You know at that time the wave will swamp your boat when it pulls the rope tight. It would be better to cut the rope and try to float on the wave. But a smart sailor knows how to use a “sea anchor,” a device that works to keep your boat still by connecting it to the surface near the boat.

Electrical power engineers know the condition well, and they call a similar electrical wave ground potential rise. The sea anchor is analogous to the alternative earthing electrode, your ground rod at the phone booth. But, because power safety prohibits us from cutting loose from the shore, the code then requires us to back up the electrode with the bonding conductor.

In fact, in the NEC section formally titled “NFPA 78--The National Lightning Code,” the alternative earthing electrode is required. If you don’t provide it and someone is hurt in or near your booth, you can be liable for some very large damage costs because you didn’t take “prudent measures” to protect people against hazards from ground potential rise.

Besides, it just makes good business sense to protect the microelectronics boards in your phone. What’s happening to those microelectronics? It takes a surgeon’s microtome and a microscope to see it. Right where the 5-volt Zener diode is on the integrated circuit chips, a sliced-open chip will reveal a microscopic volcano where the diode used to be. Why? Because one end of that diode has to be connected to what it calls “ground” to make the circuit work. The amount of current the diode has to conduct for millionths of a second when its ground floats to thousands--perhaps only hundreds--of volts from normal is so large that the Zener diode explodes. All it takes is one hit and your smart pay station never gets a chance to dial you up for its own obituary. It just dies instantly.

So why didn’t this affect pay stations before? That’s easy: the electromechanical devices of the monopoly era, those simple relays and switches, could take repeated hits before they would fail. The dissipating voltage, if high enough to puncture a relay’s varnished cotton and even plastic insulation, simply leaves a pinhole in the insulation, and if the operating voltage of 48 volts is still much lower than the general insulation value of 600 volts or so, the relay keeps working. In fact, it works until enough “pinholes,” perhaps dozens or hundreds of them finally, cause so much insulation loss that the relay gets enough shorted turns that it will not work anymore. Technicians curious enough to unwind a failed relay often wonder what causes the wire’s peppered appearance.

For this reason, many “experienced” installers who put in electromechanical pay phones for years never believed the one lesson they got in bonding and grounding. It seemed to be a theoretical thing that the teacher just wanted to use for an exam question. They had many years of practical experience that led them to believe an extra ground was not necessary.

But a ground rod is not forever, either. To do its job, it must be in contact with moist soil. That sets up a constant corrosive action that eventually forms an insulating layer of corrosion on the surface of the rod, so it loses contact with the soil. An annual maintenance check using a megger--or the more sensitive devices we now have--can show deterioration of the grounding path.

The Rural Electrification Administration charted the United States years ago to show that the entire country, except for a narrow strip at the Pacific Ocean, has more than five lightning days a year, enough to put 95 percent of the nation at risk for lightning damage. (It’s a kind of natural tradeoff for earthquakes). But, even at the Pacific shoreline, with only one day of lightning per year, one lightning bolt nearby can kill your modern microelectronic pay station.

As our good ol’ boys near Southeast lighting centers say, “You’d better get your ground down to earth, son!”

Written by Donald E. Kimberlin, principle consultant of Telecommunications Network Architects in Safety Harbor, Florida.

This article is adapted from Vol. 3, No. 1, p. 50.

Read 2500 times Last modified on Thursday, 11 October 2012 08:18