[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Fwd: [LightningProtection] Got Ground?]



Original poster: "davep by way of Terry Fritz <twftesla-at-qwest-dot-net>" <davep-at-quik-dot-com>

[This, while not directly Tesla Coil, may be of interest.
It fell of a professional lightning list.

best
 dwp]
======================================================================
-------- Original Message --------
Subject: [LightningProtection] Got Ground?
Date: Mon, 05 Nov 2001 18:54:43 -0000
From: "R.T. Hasbrouck" 

[Lightning safety discussions almost always include some reference to 
grounding. Yet, many of the comments and/or questions I see indicate 
to me that the concept of ground is often poorly understood. The 
attached document provides an introductory overview of several types 
of grounds, including lightning safety.

Rick O'Keefe has informed me that several experts reviewed and made 
no further suggestions to the write up, so I am confident it won't 
convey erroneous information.]

GROUNDING, AN OVERVIEW

R.T. Hasbrouck, EE, PE (Control Systems)
Core-group member

Although the term ground is repeatedly used, and misused, when 
discussing lightning, I suspect some don't realize that not all 
grounds are the same. The following is not a rigorous engineering 
dissertation, but an attempt to highlight the different types of 
grounds. 

A true ground is an equipotential plane, which may be the Earth 
itself (actually located somewhere beneath Earth's surface—the 
British use the term "earthed" rather than "grounded") or a large 
metallic surface. Ideally, the potential difference between a 
grounding point and true ground will be zero volts, and all points 
that are grounded will have zero potential difference between them. 
Since current will flow through grounding paths and through ground 
itself, the only way for potential differences to be zero is for the 
impedance of the path to be zero. In the real world, the best that 
can be hoped for will be low values of impedance.

Grounding for electrical power systems is quite different from a 
radio-frequency ground, an electrostatic ground, or a lightning 
ground. For electrical power systems, which operate at frequencies 
close to dc, a metallic rod (having a low value of dc resistance) 
driven into the earth works fine (refer to the National Electrical 
Code for details). This ground is often referred to as a safety 
ground. If an internal failure (fault) causes the "hot" power lead to 
come into contact with the metal housing of an appliance, tool, or 
piece of equipment, a person touching that housing will receive an 
electrical shock. The severity of the shock will depend upon the 
amount of insulation between the victim and ground. The "ground" with 
which the victim is in contact might be a wet concrete floor, 
metallic plumbing, or other grounded electrical equipment. To protect 
against such shocks, electrically powered appliances, tools, and 
equipment having metallic housings must be connected to the safety 
ground. When so connected, fault current will flow via the safety 
wire to ground, causing the metal case to be close to zero volts, 
rather than at the full line voltage (assuming the safety ground 
circuit resistance is sufficiently low). Safety ground conductors 
need to be capable of carrying the full load current of the 
associated circuit. Note that if the fault causes full load current 
to flow, a fuse or circuit breaker should trip; if the circuit has a 
ground fault interrupter (GFI), a small amount of current will trip 
the GFI.   

For equipment operating at radio frequencies, it is not always 
necessary to have a direct connection to an equipotential surface—
capacitive coupling is often used to provide a low-impedance path to 
ground for RF current. 

Lightning, which is a current pulse, contains a broad spectrum of 
frequencies—the center of the power spectrum is about 4.5 kHz, with 
the upper limit reaching into the MHz range. Its peak return-stroke 
current is extremely large (10s of thousands of amperes), typically 
lasting for a hundred microseconds, or so. As the return-stroke 
current pulse flows through the resistance of the earth it produces a 
very large transient potential gradient across the ground. This 
potentially lethal gradient—nominally 1,000 volts per meter—is known 
as step voltage. However, even when the current is flowing in a 
substantial metallic conductor (i.e., one having a very low value of 
dc resistance) very large transient voltages are developed along the 
conductor. Although resistance may be very low, e.g., less than 10 
ohms, the inductance (L) of the conductor (nominally 1.5 microhenrys 
per meter of conductor length) times the very high rate of change 
(di/dt) of the current pulse produces transient voltages reaching 
100s of thousands volts, or higher (V == I*R + L* di/dt). So, despite 
the big emphasis on achieving a very low resistance ground, the 
inductive effect predominates, resulting in transient voltages 
significantly higher than those attributed to dc resistance of the 
grounding system. A lightning grounding system must be capable of 
accommodating extremely high peak currents, and present low values of 
resistance and inductance. When grounding system resistance is 
tested, the equipment operates at a very low frequency. The result, 
which may look quite low, will actually be just the dc resistance 
component. Huge (i.e., deadly and damaging) transient voltages will 
be developed across the conductor while return-stroke current is 
flowing. Finally, consider a ten-meter section of heavy copper 
conductor connected to an earth ground. For lightning protection, two 
systems are bonded to it, one at each end of the section. The dc 
resistance between the two points is measured to be ten milliohms; 
the inductance is 15 microhenrys. A 50th percentile lightning return 
stroke of 24-25 kA, with a current rate-of rise of 40 kA/microsecond, 
flows through the conductor. Peak current times dc resistance 
produces approximately a paltry 240 V peak between the two "grounded" 
points. However, the peak transient voltage resulting from the 
conductor's inductance is 600,000 volts! The two supposedly grounded 
systems are 600 kV apart, albeit only for a brief interval of time. 
Equipment damage and serious injury or death are definite 
possibilities, hence the reason for using single-point grounding.  

Finally, consider the bonding of a fueling vehicle to a recipient 
tank. The purpose for bonding is to ensure an equipotential 
environment so that no electrostatic discharge can take place between 
the tanker and the tank. Just the process of transferring fuel can 
generate enough electrostatic potential difference to produce an 
ignition-producing spark. In another situation, workers handling 
sensitive explosives or sensitive solid state devices must not be 
allowed to develop electrostatic charges on their bodies. Static 
charge can reach multi-kilovolt-level voltages. However, unlike 
utilities and lightning, the associated currents are very small—
nominally 100s of microamperes. Consequently, the resistance of the 
discharge path can have a high value. People who work with 
electrostatically sensitive components or explosives wear metal 
watchbands bonded to the facility grounding system. To ensure that 
workers so grounded won't be electrocuted if they accidentally come 
in contact with utility power, a high-value, current-limiting 
resistor (typically 1 megohm) is placed in series with the bonding 
strap. 

In conclusion, when discussing "grounding" it is very important to 
know what type of ground is being considered, and to recognize that 
ground is not always ground.