How does one prove that the device works except by controlled lab tests? Both are reputable companies with long history in making electrical devices. Read the specifications or toss a coin.

The problem with any MOV based device (and they just about all use MOVs to clamp the surges) is that every surge damages the device, reducing it capacity to handle current the next time.
You know it has failed when it fails catastrophically.

The test to determine current handling is destructive, so it cannot be done on devices you intend to use.

The MOV arrestors work, but if you have a lot of surges you should replace them periodically.

My elecriican on the new build is not a huge fan of them, but said he will install it for no charge. I did find one that uses a Fuse approach instead of the MOV, but it's pricey. I'll guess I will just go with the Leviton since it's cheaper and he'll install for free.

"I did find one that uses a Fuse approach instead of the MOV, but it's pricey."

Fuses blow on current, not voltage.

By the time the current is high enough to cause damage, the voltage will already have done its damage.

No protector does protection. Surge protection is done best by connecting every incoming utility wire directly and short (ie 'less than 10 feet') to earth ground. Because earth does the protection. Because hundreds of thousands of joules are absorbed by earth - not by any protector.

Some incoming wires cannot connect directly to earth (ie telephone, AC electric). So a protector must make that short (ie 'less than 10 foot') connection to earth. How good is that protector? How good is the earthing? Best protection is created by upgrading earth ground. Installing it when footings are poured.

Learn about Ufer grounds. Another superior solution is to encircle the foundation with a bare copper wire. Also a superior earth ground that makes a connected protector better.

Why do so many electricians not appreciate the superior and 100 year old proven solution? Electricians are only taught what to connect to what for human safety. The electrical concepts so critical to surge protection are not taught to electricians. For example, a connection to earth must be low impedance. That means 'less than 10 feet'. No sharp wire bends. Grounding wire not inside metallic conduit. Ground wire routed separated from other non-grounding wires. Electricians would never understand the electrical significance of any of that.

For example, connect a 200 watt transmitter to a long wire antenna. Touch one part of that antenna. No voltage is felt. Touch another part to be shocked by over 100 volts. How can two different voltages be found on the same wire? Only electricians who understand basic electrical concepts could also understand why two different voltages on the same wire. Or understand that earth ground defines protection. Connection from 'whole house' protector must be low impedance (ie 'less than 10 feet').

A wire leaving a basement breaker box, going up over the foundation and down to an earth ground rod would meet safety code. And compromise the 'whole house' protector. That wire must go through the foundation and down to earth. Some reasons are in the above paragraph.

Waste no time learning why surge protection is best installed when the footing are poured. Rebar must be properly installed and bonded so that concrete footings provide surge protection.

Moving on to the protector. Its job is to connect a surge �low impedance� to earth when a wire cannot do same. A direct lightning strike is typically 20,000 amps. A minimal �whole house� protector starts at 50,000 amps. Because direct lightning strikes must even result is a functional protector.

50,000 amps define a protector�s life expectancy. Earth ground defines protection "system" effectiveness. You must address both. Latter must be addressed ASAP.

Westom, I'm sure you mean well, but let's keep these discussions to proven science. Surge suppressors do work, even if the name "suppressor" is a bit misleading. Let's agree that in a perfect world, they should be called "surge shunters". Give it a rest.

Above is the proven science. Done in every facility that cannot have damage. Again, a protector does not do protection. It only connects to what does protection. Some protection systems do not even have protectors. But every protection system has what must always exist to have surge protection. Single point earth ground.

Two completely different devices are called protectors. Popular ones cost tens or 100 times more money per protected appliance. Do not connect low impedance to earth. Another type from more responsible companies (ie Square D, Leviton, GE, ABB, Siemens, Cutler-Hammer, Intermatic, Polyphaser, Keison) actually does something useful. Makes a connection to the only item always required in every effective system. Earth ground.

Best protection, that also costs least money, exists if earthing is installed before the foundation is backfilled. Then he has the solution found in every facility that cannot ever have damage - including munitions dumps. What was pioneered in munitions dumps so that direct lightning strikes do not cause explosions? Ufer ground.

Protection is where hundreds of thousands of joules are harmlessly absorbed. How to make any protector better? Install superior earthing before a foundation is backfilled. Only that ground defines quality of protection. No earthing, then ineffective protection exists. That science was well understood and proven for over 100 years. A protector is only as effective as its earth ground.

While ufer grounds have advantages (especially in dry and poorly conducting soil) they are not normally an option for simple retrofit and in soil that is a good conduct they may be of little to no benefit.

Depending on the actual source of the surge any earth ground may be of limited benefit.

The charge in the surge is trying to t\return to its source, not return to earth.

A surge is connection to what the surge seeks - earth ground. Either a surge dissipates harmlessly outside. Or it hunts for earth destructively via appliances. Those are the only two alternatives. Once inside, nothing can stop that hunt. Is why all facilities that cannot have damage upgrade earthing as much as practicable. Military required regular inspections because earthing is that critical to electronic protection.

A ten foot earth ground rod is a significant increase in protection. Every addition to that earthing system does even less improvement. Facilities such as telcos will spend $thousands more for earthing just to make it a tiny bit better. Because protection is always defined by the earthing. As was proven by Dr Ufer when Ufer ground were pioneered in munitions dumps. Used everywhere.

Quality of earthing defines protection. Other factors that also improve protection are a shorter (lower impedance) connection to earth. And increasing distance between an earthed protector and appliances.

The best information on surges and surge protection I have seen is at:
http://www.lightningsafety.com/nlsi_lhm/IEEE_Guide.pdf
- "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 in 2005 (the IEEE is the major organization of electrical and electronic engineers in the US).
And also:
http://www.nist.gov/public_affairs/practiceguides/surgesfnl.pdf
- "NIST recommended practice guide: Surges Happen!: how to protect the appliances in your home" published by the US National Institute of Standards and Technology in 2001

The IEEE guide is aimed at those with some technical background. The NIST guide is aimed at the general public.

westom:
"A ten foot earth ground rod is a significant increase in protection. "

Nonsense.

If you have a surge to earth of 1,000A and the resistance to earth of a ground rod is a near miraculous 10 ohms, the building ground system will be at 10,000V above 'absolute' earth potential. A lot of the protection is having all wiring inside the building rise together. That requires short ground wires from entry protectors for phone and cable (and dish) to a common connection point with the power earthing system.

Most new construction in the US requires installation of a "concrete encased electrode", commonly called a Ufer ground, which is far better than ground rods.

weatom:
"Quality of earthing defines protection. Other factors that also improve protection are a shorter (lower impedance) connection to earth."

The author of the NIST surge guide has written "the impedance of the grounding system to 'true earth' is far less important than the integrity of the bonding of the various parts of the grounding system."

That is, there must be short ground connections from phone and cable entry protectors to the common connection point with the power earthing system. An example of a connection that is too long is in the IEEE surge guide starting page 30. If the connection is not short you can get high voltages between power and phone/cable wires. The NIST surge guide suggests that is what causes most equipment damage (for instance to TV related equipment connecting to both power and cable).

The maximum surge that has any reasonable probability of occurring is 10,000A on each service hot wire. (That is with a 100,000A strike to a utility pole adjacent to the house in typical urban overhead distribution.)

The IEEE surge guide recommends 20,000 to 70,000A surge current rating per wire. In areas of high lightning the recommendation is 40,000 to 120,000A. High ratings mean the protector will have a long life.

The SquareD and Leviton devices are within both suggested ratings. Length of wire from the protector to the panel connection is critical (and discussed in the IEEE surge guide). I prefer protectors that mount through a knockout in the panel (connecting wires will be shorter). For some panels you can get surge protectors that plug-in as a breaker. For SquareD QO panels it is QO2175SB with a rating of 22,500 surge amps per wire.

A version of the SquareD protector, SDSB175C, is for interior installation. Between the protectors listed I would probably use the Leviton one. If you are in a high lightning risk area the SquareD one has a higher rating (which means longer life). If telephone or cable entry points are distant from the power service point the SquareD one has ports where the cable and phone wire through the protector.

"A surge is connection to what the surge seeks - earth ground."
So when I use my surge generator (1 microsecond/1000 microsecond/2,000 V) it is trying to reach earth?

The voltage is generated on the secondary of a transformer, and is seeking the other side of the transformer winding, no other place, and certainly not earth ground.

It will operate to specification into any load of more than 5 ohms impedance across the output.

Surges are always seeking the source that created the current flow, not earth unless it was already involved in generating the surge.

Short a 24 kV power distribution line to a 7.2 kV distribution line.

The surge created on the lower voltage line is seeking to return to the transformer or generator that created the 24 kV difference in potential.

It will use the earth as a conductor to return to the source if the source is earth referenced.

When a magnetic field induces current to flow in a conductor, the current is seeking to return to the place it left, the wire (or in a constantly changing field the other end of the wire it left from).

in many soils ufer grounds are gross overkill.

There development was driven by poor ground connections in dry soil.

The concrete acts as an ion donor to increase the conductivity of the soil when moisture is present, ad the ions remain when the moisture departs.

If the soil is damp and remains damp there are normally more than enough ions around for good conduction, allowing simple rod electrodes that make contact with the moist soil to be effective grounds.

Ufer grounds can easily be created during the construction of reenforced concrete buildings by ensuring bonding of the reinforcement bars electrically (often by welding them together).
When they are buried in concrete you create a very effective earth ground.

It is normally more than deep enough in all but the driest soils to have good water table contact.
In those areas the foundation can actually be 'watered' to improve earth bonding.
A perimeter drain system can be blocked at its outlet and flooded.
It does run the risk of water intrusion into the building.

The 'surge suppressor' devices (MOVs being a very common type for this) allow for a connection between wiring and ground/neutral when the voltage exceeds their break-over rating.

This has the effect of clamping the voltage on the line to approximately the break over voltage of the device.
If the surge current is not very large or very brief the power dissipated in the device (current times break-over voltage times time conducting) will not cause destruction of the device

For MOVs it does cause some localized damage that decreases the ability of the device to carry current the next time it breaks over into conduction.

In the most common description the device is said to 'suppress' or 'shunt' the surge to ground.

A poor connection to the source of the surge can result in limited effectiveness, and that includes the path back to earth ground for surges referenced to earth ground.
The impedance of the ground path ads to the voltage the clamping device needs to turn on, decreasing its effectiveness at liming the voltage rise.

Since ground rods are allowed to be as high as 25 ohms to earth, this can create a significant voltage drop if the surge current is large (a relatively uncommon occurrence for the most part, unless you have lightning involved or high voltage POCO distribution faults).

Most rods are far lower than even 25 ohms reducing the effect (though long ground leads can increase the impedance of the circuit, especially to quickly varying voltages).

The typical 'static spark' in the winter can be many tens of thousands of volts (and is more or less earth seeking most of the time) but us VERY current limited.

In areas with significant lightning possibility the suppressors can be a worthwhile investment, just make sure they are well grounded to earth (low impedance, not just resistance) and periodically replaced.

The TELCOs have spent many years and may dollars trying to reduce the damage to their equipment, connected to your phone my long lengths of phone lines often sting on poles.

Even proximity to a lightning strike induce large voltage in the lines, and they are nicely carried back to the central office equipment (now often semiconductor electronic based) and also to your phone (or modem, or fax, or anything else connected to eh phone line in your house).

Through all this the protection must deal withe -48 VDC across the wire pair, ad 90 VAC 20 Hz ringer voltages that are impressed on top of the -48 VDC, and the nominal 1 VAC of the actual audio signals.

They are pretty good at protection though, ad use a cascade of devices with smaller and smaller break-over voltages to limit what finally is presented to the semiconductor equipment.

Luckily the insulation on the phone lines usually fails around 4,000 to 6,000 volts (line to line of each pair) providing at least an upper limit on what is transmitted for many distances.
Of course the line could still be hit right outside the central office though.

A 33kv wire fell on local distribution. Electric meters literally exploded up to 30 feet from their pans. Many - especially those with plug-in protectors - had damage even to their 'profit center' protectors.

My friend knows someone who actually does this stuff. My friend properly earthed one 'whole house' protector. Had no damage except to his electric meter. Nothing new. Protection has been done this way for over 100 years.

A protector is only as effective as its earth ground - as my friend learned when 33,000 volts threatened and could not damage any appliances.

Surges (ie lightning) may be 20,000 amps. A minimally sized 'whole house' protector starts at 50,000 amps to make all surges irrelevant. Connects hundreds of thousands of joules harmlessly to earth. A protector is only as effective as its earth ground - as defined by numbers.

"A protector is only as effective as its earth ground - as defined by numbers. "

Only for earth referred surges.

brickeyee wrote:
"Ufer grounds can easily be created during the construction of reenforced concrete buildings by ensuring bonding of the reinforcement bars electrically (often by welding them together).
When they are buried in concrete you create a very effective earth ground."

Under the US National Electrical Code a "concrete encased electrode" is usually required for new construction. The required minimum length is 20 ft and can be 1/2" rebar tie-wired together where necessary. I agree they are good. Ground rods are lousy.

brickeyee wrote:
"The 'surge suppressor' devices (MOVs being a very common type for this) allow for a connection between wiring and ground/neutral when the voltage exceeds their break-over rating.
This has the effect of clamping the voltage on the line to approximately the break over voltage of the device. If the surge current is not very large or very brief the power dissipated in the device (current times break-over voltage times time conducting) will not cause destruction of the device"

Maximum service surge current is 10,000A per service wire. Service panel protectors with much higher ratings are readily available.

The author of the NIST surge guide looked at the amount of energy that might actually be absorbed in a plug-in protector. Branch circuits were 10m and longer. Power line surges were up to 10,000A (as above). The maximum was a surprisingly small 35 joules (based on US wiring). Plug-in protectors with far higher ratings are readily available.

brickeyee wrote:
"For MOVs it does cause some localized damage that decreases the ability of the device to carry current the next time it breaks over into conduction."

As MOVs absorb surge energy they eventually start to conduct at lower voltages. Eventually they conduct at normal voltages, and go into thermal runaway. (Since 1998 UL has required listed protectors have thermal disconnects to disconnect overheating MOVs.) Far as I know MOVs continue to conduct surge currents up to failure.

(Protection is not by absorbing the surge, but some energy is absorbed in the process of protection.)

brickeyee wrote:
"In the most common description the device is said to 'suppress' or 'shunt' the surge to ground.

A poor connection to the source of the surge can result in limited effectiveness, and that includes the path back to earth ground for surges referenced to earth ground.
The impedance of the ground path ads to the voltage the clamping device needs to turn on, decreasing its effectiveness at liming the voltage rise."

For a power service protector the MOVs are connected line-to-line and line-to-neutral. In US services the neutral is bonded to ground and the earthing electrodes at the service. Unless the wires connecting the MOVs are too long they will function as intended, limiting the voltage between hots and neutral. If the earthing path is higher impedance it means that the power wiring will rise to a higher voltage above 'absolute' earth potential. With significant surge earth current it will rise in any case. Much of the protection is that all wires - power, phone, cable, ... - rise together as explained in another post.

The same happens at plug-in protectors. All wires to a set of protected equipment has to go through the protector. The protector limits the voltage on all wires to the ground at the protector. As explained in the IEEE surge guide (starting page 30) the voltage at the protector may rise well above the ground at the service but the voltages between wires to the protected equipment is safe.

brickeyee wrote:
"Luckily the insulation on the phone lines usually fails around 4,000 to 6,000 volts (line to line of each pair) providing at least an upper limit on what is transmitted for many distances."

Interesting - I never saw a breakdown number before.

In US services, when service panel busbars reach about 6,000V there is arc-over to the enclosure. After the arc has stabilized the voltage is hundreds of volts. (Since the enclosure is connected to the earthing electrodes that dumps most of the surge energy to earth.) Similarly, 15 and 20A US receptacles will arc-over at about 6,000V.

I generally agree with brickeyee.

==========================
westom wrote
"A 33kv wire fell on local distribution. Electric meters literally exploded up to 30 feet from their pans.
My friend properly earthed one 'whole house' protector. Had no damage except to his electric meter."

MOVs can conduct thousands of amps for the maybe 100 microseconds of a lightning surge. They will be rapidly destroyed by crossed power lines. No protector will say it can protect from crossed power lines. (UPSs and a few plug-in protectors disconnect on over-voltage and may provide protection.)

westom wrote
"Surges (ie lightning) may be 20,000 amps."

The maximum surge current that has any reasonable probability of occurring is 10,000A per service wire - worst case. There is a reference to it in the IEEE surge guide.

westom wrote
"A protector is only as effective as its earth ground"

Next worst surges to lightning are normal and abnormal utility operations including fault clearing and switching power factor correction capacitors. Many of them are line to neutral, not earth. I agree with brickeyee.

Airplanes are regularly hit by lightning and have no earth connection. Understanding the protection is not that difficult.

"Under the US National Electrical Code a "concrete encased electrode" is usually required for new construction."

Nope.

If you have a metal water line with 10 feet in contact woth earth it is required to be the promary electrode.
It must be sumplemented by a single rod (10 feet long).

If you donot have a water line that meets the requirements you canuse a single rod, but it must be less than 25 ohms top earth.
A second rod is then required if it is over 25 ohms.
Since measuring the first rod requires a second rod be driven (and then removed) adding the second rod from the start is common.
It hen removes any requirement to measure the resistance to ground.
Ufer grounds, plate grounds, wire grounds, etc. are also acceptable as made electrodes.
In a poured concrete building making a ufer ground is simple and cost effective )it just takes some rebar bonding).
In a residential setting it is not since the slabs do not normally have much rebar, just wire mesh for crack control.

> If you have a metal water line with 10 feet in
> contact woth earth it is required to be the
> promary electrode. It must be sumplemented by a
> single rod (10 feet long).

Water pipe earth ground is the only electrode so insufficient as to require supplemental earthing. Any other earthing electrode is sufficient. But water pipe earth ground requires some other earthing.

McGraw-Hill's "National Electrical Code Handbook" (their capital letters):
> EVEN THOUGH THE WATER PIPE IS A SUTIABLE GROUNDING
> ELECTRODE ... AT LEAST ONE MORE GROUND ELECTRODE
> MUST BE PROVIDED .... A water pipe, by itself, is
> not an adequate grounding electrode to provide a
> "grounding electrode system".

Encircling a building is even important in barns. Otherwise milk production can be diminished.

That requirement defines human safety. OP's earthing must both meet and exceed safety code requirements. An Ufer ground or to encircle the building with a buried copper wire.

Others have used variations of this superior solution. An example of why the radio station does not suffer damage:
http://scott-inc.com/html/ufer.htm

"Others have used variations of this superior solution."

A ufer ground is often more effective, as is a lerger electrode if it is low inductance.

The simple wire loop type often fails on the inductance side of things.

Wires tend to look more and more inductive as the frequency of the EMI rises.

The moderately fast rise time of lightning is usually not to bad, but other signals can have faster rise times.

A ufer is often overkill for a residential grounding electrode system, and expensive.

The ufer will not add much protection at all unless the grounding electrode conductors are suitably low impedance at the signal rise time of interest.

Real EMI type work often makes use of flat braid to reduce the inductance of the conductor, and they can be rather wide and large (4-6 inches wide) to achieve the low impedance required.

bud wrote:
"Under the US National Electrical Code a "concrete encased electrode" is usually required for new construction."

brickeyee wrote:
"Nope."

quote
NEC 250.50 Grounding Electrode System. All grounding electrodes as described in 250.52(A)(1) through (A)(7) that are present at each building or structure served shall be bonded together to form the grounding electrode system.
...
Exception: Concrete-encased electrodes of existing buildings or structures shall not be required to be part of the grounding electrode system where the steel reinforcing bars or rods are not accessible for use without disturbing the concrete.
...
250.52(A)(1) Metal Underground Water Pipe ....
...
250.52(A)(3) Concrete Encased Electrode ...
/quote
Not the clearest writing, but from the exception "concrete encased electrodes" are required for new construction when the construction method is compatible.

brickeyee wrote:
"If you have a metal water line with 10 feet in contact woth earth it is required to be the promary electrode."

There is no "primary" electrode. There is an "electrode system" which may (or may not ) have more than one component. In the case of water pipe, there is a required "supplemental" electrode.

brickeyee wrote:
"It must be sumplemented by a single rod (10 feet long)."

Because metal water pipe may be replaced by plastic it must have a supplemental electrode that can be one of several types. One is a ground rod (8 ft minimum); another is a "concrete encased electrode".

brickeyee wrote:
"If you donot have a water line that meets the requirements you canuse a single rod, but it must be less than 25 ohms top earth.
A second rod is then required if it is over 25 ohms.
Since measuring the first rod requires a second rod be driven (and then removed) adding the second rod from the start is common.
It hen removes any requirement to measure the resistance to ground."

I agree 2 rods are a lot simpler (if you use rods).

brickeyee wrote:
"Ufer grounds, plate grounds, wire grounds, etc. are also acceptable as made electrodes.
In a poured concrete building making a ufer ground is simple and cost effective )it just takes some rebar bonding).
In a residential setting it is not since the slabs do not normally have much rebar, just wire mesh for crack control."

If you have a concrete foundation or footing you likely need a "concrete encased electrode". In the frost belt you need footings. Don't know about a pure slab, but I thought the UBC required a footing for everything (probably some other building codes are still in effect). It does not have to be rebar - it can also be #4 bare copper.

The NEC does NOT require a concrete encased electrode, as was posted.
I did not say you could not use one, just that it i not required.

A water pipe has been the required primary electrode (if it is present and at least ten feet long) for many years and remains a common primary electrode.
The supplement was added in the 1970s and had nothing to do with plastic pipes, but it had been recognized that the single water pipe electrode was not always working as desired.
Occasionally the metal pipes would have enough voltage and current on them to be a hazard when utilities worked on the water mains.

westom wrote:
"Water pipe earth ground is the only electrode so insufficient as to require supplemental earthing. Any other earthing electrode is sufficient. But water pipe earth ground requires some other earthing."

Water pipes are, of course, a good earthing electrode. The problem is that they may in the future be replaced by plastic.

Does westom agree with what he has written previously?
"Water pipe is no longer acceptable as an earth ground"
"Electric wire connections to water pipes are permitted only to remove dangerous electric currents from those pipes."
"In short, any connection to pipe must be only to remove electricity - not dump electricity into
a pipe."
The NEC requires "earthing electrode that conforms to paragraphs two through seven." (Water pipes are paragraph one.)

Simple yes or no answer - are water service pipes (min 10 ft metal in the earth) REQUIRED to be used as an earthing electrode?

westom wrote:
"Encircling a building is even important in barns. Otherwise milk production can be diminished."

There is an article (547) on "Agricultural Buildings" with extensive bonding requirements (kind of like swimming pools).

westom wrote:
"Others have used variations of this superior solution. An example of why the radio station does not suffer damage:
http://scott-inc.com/html/ufer.htm";

Good comments by brickeyee on this.

Yes, if you are going to put a commercial broadcast tower in you backyard (guaranteed to be hit by lightning) it is a good reference.

The "concrete encased electrode" in the NEC is far less extensive than the "Ufer ground" that was investigated. "Concrete encased electrodes" however are a good earthing electrode.

brickeyee wrote:
"The NEC does NOT require a concrete encased electrode, as was posted.
I did not say you could not use one, just that it i not required."

My original language was "usually required".

If you have a concrete footing or foundation an inspector will want a "concrete encased electrode". That is virtually everything here in the frost belt. Enforceable sections of the NEC were provided.

If you have a slab without a footing you may not need one - as I tried to say I am not familiar with building codes for them.

brickeyee wrote:
"A water pipe has been the required primary electrode (if it is present and at least ten feet long) for many years and remains a common primary electrode.
The supplement was added in the 1970s and had nothing to do with plastic pipes, but it had been recognized that the single water pipe electrode was not always working as desired.
Occasionally the metal pipes would have enough voltage and current on them to be a hazard when utilities worked on the water mains."

I have not heard of problems with utility workers.

My oldest NEC, 1968, has:
" ... or there is some likelihood of the piping system being disconnected or isolated through the use of nonmetalic piping or insulated couplings, the piping system shall be supplemented by one or more of the grounding electrodes recognized in ...."

Later (1978) as plastic water supply systems became more common all new water pipe electrodes were required to have a "supplemental" electrode.

From the 1996 National Electrical Code Handbook - published by the NFPA:
"The requirement to supplement the metal water pipe is based on the practice of using plastic pipe for replacement when the original metal water pipe fails. This leaves the system without a grounding electrode unless a supplementary electrode is provided."

-------------------------------
I agree most of what you have written in this thread

Hello! I think you can use Phoenix overvoltage protectors. They are installed in our house and I have never had any problems with voltage spikes. I don't remember about their price but you can easily find them in the Internet (here for example http://electrical-components.com/phoenix)