I realize this is an older thread but it seems to me that all of the participants are not considering impedance but only resistance. Since a transient is basically a high frequency phenomenon we cannot truly analyze it if we only consider resistance. The voltage imposed on the equipment is expreseed by the equation "E" equals L times "di divided by dt". Thus the inductance times the instantaneous rate of change of current with respect to time. For #6 copper wire this is approximately 4.337 microhenries for 3 meters. If the rate of change were 1000A/microsecond than the voltage drop across 3 meters of #6 AWG copper wire would be approximately 4300 volts. As an impulse moves into a home or facility the di/dt is reduced due to the series inductance and the parallel capacitance inherent in the wiring. Service entrance protection will reduce the effects imposed upon the wiring and electrical/electronic components installed.

Ultimately protection depends on careful system design since incorrectly positioned protectors may contribute to voltage differentials between ports of the equipment. These ports include power, LAN, Telco, USB etal. These differentials are what typically damage small signal transmitters and receivers. Surge reference equalizers as discussed in IEEE 1100 are the key to preventing these damages or disturances.

Edward F (Ed) Roberts - E. F. Roberts & Assoc. www.efrobertsassoc.com

I have been building a cabin in Colorado. I have a permanent 200 amp pedistal service outside with passthrough lugs to go to the house to set my 200amp panel inside. I have a couple of questions. Do I have to use 2-0 or 3-0 copper wire to connect. Also is it better to ground each seperatly. I think code requires bounding the panels together but I will be about 100' from each panel. Thanks for any advise.

I took the test for a state license i did not find the aswer in the NEC to Lenght cord for; industrial luminaire, and office (no extension cord) use for ilumination , Can you give a t

i found 6ft lenght for a city code but no intheNEC

max

Mike,

Your thought that a multigrounded primary neutral would present almost no impedance to a lightning strike is not correct.

Because of the inductance of the wire and wave phenomenon a lightning strike will only go to the 2 to 4 closest grounding electrodes.

Also, Cleveland Electric Illumination Company as well as some upstate New York utilites have ungrounded 4,400 volt and 4,800 volt distribution respectively. Part of Ohio Edison's territory out by Norwalk, Ohio has 7,200 volt ungrounded. On these systems a building theoretically needs 3 grounding electrodes plus the electrode at the utility transformer to make a modern telephone protector block ( surge arrestor ) legal.

In climates that have periodic droughts a ground rod that goes deep into the earth, such as a rod that is driven beneath a basement floor, will outperform an 8 foot rod that is driven from the surface. One of the top electrical inspectors around here says that the top 6 feet of a rod that is driven from the surface does not act as a grounding electrode because the soil is not compact enough or wet enough particularly near a building foundation.

Over at www.eng-tips.com somebody who is with a Colorado electrical utility says that on the portions of their system that have 17 foot ground rods they have zero lighting damage.

Same message as recent post - just revised format for readability

I am glad to see our little video has generated some good discussion on the important topic of surge protection.

I would like to offer the following clarifications and comments:

1). The primary purpose of the 2 minute and 2 second video was to illustrate to the layman homeowner the multi-stage approach of TVSS application as recommended by the IEEE Emerald book (IEEE Std1100-1999, 8.6.3 Large Surge Suppression). i.e. first diversion of large surge currents at service entrance, any residual voltage dealt with by a second protective device at the critical load.

2). The illustrations in the video assumed a properly grounded and bonded installation and described how current will flow under a large surge event such as a nearby lightning strike.

3). I am deeply concerned that someone might read the comment that “A common misconception I once had was that a TVSS requires a good ground to operate” to mean that a TVSS will operate correctly when connected to an improperly grounded and bonded system. Nothing could be further from the truth!! It is absolutely critical that proper grounding and bonding methods are followed when applying a TVSS. For example, one of the most common reasons for “failure” of TVSS downstream from the service entrance is the absence of a system bonding jumper on separately derived systems [250.30(A)1]. Perhaps the key point here is to distinguish the subtleties surrounding the words “good ground”. We try to be careful and specific when we emphasize in our training and literature that it is important to have a “properly grounded and bonded” system in order for a TVSS system to work correctly. I believe that this often gets interpreted and simplified to “it is important to have a good ground”. The actual impact of the contact resistance of the earth of the GES is much less important than proper grounding and bonding methods as outlined in the National Electrical Code.

4). Some of the discussion and comments on this subject seemed to imply that the only purpose of TVSS is to protect from induced transients from nearby lightning strikes. While protecting from induced transients from lightning at the service entrance is one of the functions of a TVSS, numerous studies have shown that many more transients are generated from load switching and other internal transient generating sources. The addition of L-G protection and N-G protection in TVSS is typically designed to be applied at devices deeper inside the facility, not at the service entrance.

Carey Mossop Product Line Manager - Eaton Corp.

Mike,

The performance and behavior of the Transient Voltage Surge Suppressor (TVSS) and/or the Surge Protective Device (SPD) has truly been characterized, by many manufacturers, as "magic". Lightning itself is perceived as a form of "black magic" by many individuals who have witnessed its power and unpredictability. Many individuals who have used surge protection in their homes have noticed less failures of their electrical equipment from light bulbs to expensive electronics. All of this combined has allowed many entrepreneurs to "cash in" on these and other misconceptions.

The problem is that lightning and man-made transients are just that, transients. Transient behavior can not be perceived in the same way we perceive steady state power; so grounding, circuit impedances, and other “normal" circuit parameters do not act the way we naturally expect them to behave. The transient is a relativistic, three-dimensional phenomenon that often appears not to adhere to Ohm's Law. But, with a little understanding we can predict how transients will affect our power systems and, yes, transients STRICTLY obey Ohm's Law.

First, there are several rules that we must accept if we are to successfully deal with transients. 1 - There is no such thing as a "singular ground". 2 - EVERY conductor has significant impedance regardless of its size. This includes the earth itself. 3 - Very small parametric values can be JUST AS IMPORTANT as very large parametric values. For example, we understand that kilovolts and kiloamps are large and potentially deadly. But in the transient world milli-henrys can be just as dangerous. 4- It is important to consider what initiated the transient, where the transient is coming from, and where the transient wants to go. Lightning transients must equalize the charge between cloud and earth, locally developed transients (for example a motor turning on and off) must dissipate the field energy within a local loop. 5 – Ground or conductor resistance is not enough; in the transient world reactance must be considered in addition to the DC, or low frequency, ground.

For the purpose of brevity, I am going to limit my comment to the topic of Residential Grounding as it relates to surges.

Let's start at the utility phase wire on top of the pole and for simplicities sake, let's imagine a single phase line that connects to a utility transformer and a utility high voltage surge arrester. A bolt of lightning strikes the phase wire, immediately the current divides in half. One half of the surge heads back to the utility, in search of an earth ground, and the other half proceeds to the transformer feeding the home, again looking for earth ground. The utility surge arrester reacts to the surge by diverting much of the surge current to earth. The surge arrester develops a discharge voltage based on the surge current (maybe 25,000 volts). At this point two things happen; first this "discharge voltage" is developed across the arrester is presented to the transformer to "step-down" to send to the home. Secondly, a very large surge current is sent down the utility ground wire causing a rise in the local voltage at the utility ground/neutral connection with respect to the home and the next pole span.

Let's take a moment and consider how high (in volts) the neutral/ground connection has risen at the base of the utility transformer. If there is 10 ohm ground impedance at the utility transformer and there was a 1500 amp surge diverted by the arrester, then the neutral/ground voltage would be 15,000 volts. (E=IR) The numbers for ground impedance, surge current, arrester discharge voltage, etc are rather arbitrary and arguable.

At the home then, the 120 volt line (170 volts peak) has risen to at least 400 volts (25,000 volts (discharge voltage) / 60 turns (transformer turns ratio)) and the ground/neutral (at the home) has also risen to a value between 0 and 15,000 volts based on the neutral impedance of the neutral wire and the ground impedance at the home. So at the home, the L1 & L2 to Neutral is 400+ volts, the ground to "far ground" is between 0 and 15,000 volts, the L1 to "far ground" 0 to 20,000 volts peak. It is important to note that the likely neutral/ground point, at the home, is going to be considerable less than 15,000 volts peak since grounding at the home and the impedances connecting the utility transformer to the home will probably split this voltage at in half.

At this point a couple of observations and comments: 1 - Since the home "rides" on the ground/neutral plane, Mike is correct in saying that the value of the ground is not important for proper operation of the TVSS. It will clamp the 400 volts regardless of the ground condition. 2 - That being said, if an outbuilding, pool, or spa has its own ground (intentionally or otherwise), then there can be a significant transient voltage between Line and spa ground creating a hazard. This indicates the need for additional surge protection at the spa. 3 - The lower the ground resistance at the service entrance the lower the ground rise will be at the home. The TVSS (SPD) still does its job and protects line connected equipment even with poor grounds. 4 - UL SVR ratings are the maximum clamp values, for that product. Actual SVR values can be less. UL SVR test values are based on a 500 amp surge. 5 – The transient voltages seem, and in fact are, quite high. This is why one will notice small weld (or burn) marks around the home on electrical equipment enclosures in high lightning areas like Tampa, FL. A high transient causes a spark-over which results in power current being drawn until the circuit breaker trips or the arc clears itself. 6 – The Neutral to Ground bond at the service entrance is extremely important to provide a common reference point for the rest of the home. Without this bond, the difference in between the power line neutral and the Earth ground at the home could rise to thousands of volts. This would be very disruptive to every piece of line connected line connected equipment in the home.

Other connections to the outside world also can cause problems in the home. Telephone and Cable TV connections are notorious for providing a different potential for “far ground” In these cases, there can be Line to Cable TV Ground arcs and/or large currents flowing between the Outlet (neutral or ground) to Cable TV Ground. This is a good reason to bond other services together at the service entrance.

I think it should be said that the Cutler-Hammer (Eaton) video makes no mention of the magnitude of the ground impedance. Just that surges should be diverted as soon as possible before they have the chance to travel, unabated, through the house. This is a philosophy that is consistent with the IEEE C62 and UL1449.

Thomas C Hartman TVSS Technology Manager Eaton Corporation

Is a low resistance ground important in the overall performance of SPD's? Yea or nay, depends on who you talk to. Has always been this way and probably always will. What is unanamous is the importance of a low restance ground for that which it was intended... personal safety.

If you do have a low resistance ground that you are comfortable with today will it exist next week, next month or next year? You really have no pratical way of knowing.

Mike, I would like to add my comments on this very interesting 'can of worms'. Through my years as a surge protection design engineer, the understanding of what role the ground and the grounding conductor should play or not play has seemed to remain a point of contention. Not only among engineers, but between engineering and marketing. There are engineers who feel that ground is essential to suppressing surges, and then there are marketers that love to brag about all the modes of protection that their gadget protects.

The CH video, to me, is pointing out an important concept, that surge protection works best in stages. A surge protector at the entrance is not going to catch all surges (their may be other entrances to the home). The ones that are caught may have enough let-thru to still create dangerous voltages to the down stream components. A point of use product can be used to further reduce this risk. As for the reference to ground, this was mentioned in simplification (I think), but we don't know their philosophy about this from the comment.

The service entrance is a special consideration in regards to ground because it is the one place neutral and the grounding wire become one as they are bonded together. Now, whether or not this connection becomes one with earth ground and is useful or hurtful in surge protection is another question. As Mr. Pine Brumett points out, most people have no way of knowing the quality of their earth ground connection.

If this earth ground is a solid low impedance (25 ohm) connection, it would still be a high impedance in comparison to the copper wire back to the transformer. (see Math below). If this were a zero ohm connection, then the induced lightning through the earth ground would be a much bigger problem. Each time lightning hit earth ground (no matter where) it would cause a surge in my home.

For lightning protection, we are concerned with protecting equipment, so if we look at the problem from the perspective of the equipment – it becomes a simpler task. We become only concerned with clamping the voltage at the input power to the device we mean to protect. Where the diverted current came from or goes – we don't really care.

At the service entrance, if we focus our protective devices on the job of clamping the voltage from Line to Neutral, then we will save equipment. Notice, this means 2 modes of protection (Line1 to Neutral and Line2 to Neutral). Line1 to Line 2 will be protected by the other two modes in series. Some may say that discrete protection is better, but more protection can be placed in the same space if the dielectric issues (spacing) of multiple modes is decreased. Maximizing protection for the two modes will also provide maximum protection for L1 to L2.

In regards to the level of protection needed, the specsmanship used by various manufacturers confuse this issue. This is really topic for another paper, but please do not introduce new terms such as KVA ratings when the ratings we have today are confusing enough for the customer.

Math, per Mike's request; 1.Ten feet of 6AWG with a 25 ohm ground rod; the impedance of the wire is about 2.78 ohms at a 100kHz frequency (see note below). Total impedance = 27.78 ohms. 2.50 feet of 3/0 AWG with 3 ten-foot ground rods (5 ohms combined); the impedance of the 50 feet of wire would be about 15 ohms. Total = 20 ohms.

Note: The most common surge waveforms used by UL1449 is the 8x20us impulse (8us rise time is near 100kHz) and the 100kHz ring wave.

PS; I have a work in progress about the role ground takes, or should not take, in surge protection

Great article, however they negleated to mention that lighting can also enter thru the telephone lines and the cable system as well........ These need to be protected as well.

Although lightning occasionally strikes the electrical distribution system requiring the SPD to conduct it to ground, the much more common occurance is for lightning to strike a grounded object in close proximity to the home or building. This creates a momentary potential rise to the entire ground plane including the grounding electrode conductor and neutral because of the bond at the service entrance. This can cause a large potential difference between the phase conductors and the neutral and ground. The SPD at the service entrance brings all conductors to the same potential until the lightning strike energy dissipates into the ground.

Just a comment regarding this statement: ". . A low resistive path to a low resistive earth connection and neutral point, such as a neutral conductor to the secondary utility transformer. Because the secondary neutral at the utility transformer is bonded to the primary neutral and the primary neutral is grounded to the earth at thousands if not millions of locations, there is practically no contact resistance at all."

This would only be true if the conductors between these many connections to "earth" had zero resistance and zero impedance. Such is obviously not the case. The distances and conductor sizes to the utility transformer ground, and the paths to "True Earth Ground" from that connection will all have impedance, and resulting voltage rises.

If we had no resistance losses on the power grid ground, (with it's " thousands if not millions" of connections to earth) there would be no potential for swimming pool or cow milking machine touch voltage shocks.

This whole topic quickly get me in over my head ... perhaps, if I put my concerns in a specific question, someone could answer it.

My home, built in 1940, has no ground wire. I have put nearly everything on GFCI-protected circuits, so most receptacles are of the three-prong type.

At my computer is a power strip, with built-in surge suppression.

Is this surge suppressor able to operate without a ground path?

I have a "stupid" question about ground rods. I am a service technician in New Jersey and Township officials give us greef about two things one being the primary grounding electrode to the main water pipe, building steal or ufier ground. What happens when there is no primary grounding electrode such as a plastic water pipe, no building steal and no ufier ground? Secondly why are some township officals making us use #4 thhn toour secondary grounding electrode, the ground rod, when it can only hold up to a maxium load of up to 60 amps and not just running a #6 to them for a 150- 1200 amp service replacement??? Thank you Ryan