whole house surge protectors

[John Grabowski]

*The surge suppressor will blow up when it is not properly grounded. It is designed to absorb the surge and then dissipate it to ground. With no ground it just overloads, but still provides some surge protection though it is only once. Proper surge protection begins with good grounding.

Siemens has a comprehensive online training manual that explains the process in simple terms and also shows their products. Here's an excerpt: "The clamping voltage rating of an MOV is greater than the normal supply voltage. Therefore, when a surge occurs and the clamping voltage rating of the MOV is exceeded, the MOV switches from a high resistance path to a low resistance path and excess energy passes through the MOV to ground, bypassing the connected load".

You can read the whole instructional manual here:

[trader4]

Interesting, I didn't know that connectors like the ILSCO KUPLER IPC

4/0-#6 that taps into the service conductors existed or were allowed. Some questions:

I've never used a lug kit. I take it that goes on the panel some place that does not take up a breaker slot and is typically used to then feed a subpanel? So, in this case they use it to feed the SS.

If you use the Kupler approach and tap the service conductors before the main breaker, you then have the SS connected directly without any breaker. The instructions say this is OK, but does it depend on where the SS is physically mounted? Say I mount it next to the panel, as is frequently done. Then I have it and the wires running between that Kupler and the SS with no breaker protection, no disconnect, including no main breaker. Is that allowed? I sure would not do that in my house when there are other ways to do it.

[trader4]

Just a clarification. The surge protector doesn't absorb the surge, it just shunts it to ground, assuming it has a ground. You have that process described below.

With no

With no ground at all, it would seem all it could do would be to clamp the various conductors so that the voltage on any of them relative to the other would be 600V or so.

Proper surge protection begins with good grounding.

[Frank]

I am speaking as a novice with these minor experiences. My neighbor is an elderly widow that has to rely on others.

We did have a big surge hit the house where high tension wire was shorted to house wires. It knocked out all my surge protectors and a microwave oven that was on a protector that had previously failed. Cost me about $90 for new protectors as once light is out, they are gone. Don't know about whole house protector reset or replacement in an incident like this.

My brother had lightening hit his driveway light pole. Killed everything in the house. Don't think anything would have protected him.

[John Grabowski]

Just a clarification. The surge protector doesn't absorb the surge, it just shunts it to ground, assuming it has a ground. You have that process described below.

*Thanks for the clarification Trader4. You are correct about the shunting.

With no

With no ground at all, it would seem all it could do would be to clamp the various conductors so that the voltage on any of them relative to the other would be 600V or so.

*I'm not sure about that. All I know is that the ones that blow up are usually the ones with no ground.

Proper surge protection begins with good grounding.

[John Grabowski]

Interesting, I didn't know that connectors like the ILSCO KUPLER IPC

4/0-#6 that taps into the service conductors existed or were allowed. Some questions:

I've never used a lug kit. I take it that goes on the panel some place that does not take up a breaker slot and is typically used to then feed a subpanel? So, in this case they use it to feed the SS.

If you use the Kupler approach and tap the service conductors before the main breaker, you then have the SS connected directly without any breaker. The instructions say this is OK, but does it depend on where the SS is physically mounted? Say I mount it next to the panel, as is frequently done. Then I have it and the wires running between that Kupler and the SS with no breaker protection, no disconnect, including no main breaker. Is that allowed? I sure would not do that in my house when there are other ways to do it.

*I agree with Trader4. I don't like having anything connected in a home that cannot be shut off without pulling the electric meter.

[trader4]

That's interesting. I take your word for it. It's just that I'm having a hard time figuring out why that would happen. With a proper ground almost all the surge energy passes directly through the surge protector. Without it, it would seem a lot less energy would pass through it. Maybe Bud or one of the other guys has some thoughts.

[bud--]

Nice description - I agree with all of it. I would only add what is implied - for a plug-in protector connect all interconnected equipment to the same protector and all external wires (power, phone, cable, ...) go through the protector.

And more good information. Many of the major electrical manufacturers make service panel protectors. I would buy only a major name brand.

------------------------- Excellent information on surges and surge protection is at:

The recommended surge current rating for residential is 20,000 - 70,000A per service wire. For high lightning areas the recommendation is 40,000

- 120,000A.

An investigation of the possible surge on a residential service used a

100,000A strike to the near utility pole with typical urban overhead distribution. Only 5% of lightning strikes are stronger, and the strike is about as close as possible, so this is essentially a worst case. The surge current was 10,000A per service wire. Higher protector ratings are not for a single event, but give a long life.

[bud--]

Most service panel protectors must be connected after the service breaker. I would rather not connect ahead of that.

--------------------------------- The protectors that a utility will provide (usually lease) connect between the meter and the meter base.

[bud--]

The protection in almost all surge protectors are MOVs. They can conduct thousands of amps for the maybe 100 microseconds of a lighting created surge.

They will be rapidly burned out by the much longer overvoltage of a crossed power wire (which is not a "surge"). Neither service panel or plug-in protectors will give reliable protection. UPSs might disconnect and some plug-in protectors disconnect on overvoltage.

[bud--]

Several manufacturers make surge protectors that plug-in like a circuit breaker - SquareD comes to mind. (They only can be used in panels made by that manufacturer.)

They can only be used at the service panel, because the neutral and ground are bonded in the service panel. (At a subpanel the protector connects to hot-hot-neutral-ground.)

Separate protectors are available with higher surge amp ratings. But the plug-onto-the-busbar ones are sure convenient.

[bud--]

IMHO the service panel protector would not have a problem with grounds that are not as good. Since a lightning surge is a current source, the current through the protectors might be about the same, but not more. (But in about all cases there are multiple paths to earth, which would result in a lower surge current.)

Suppose you threw all your beer bottles in the back yard then built a metal shack on top of the bottles. Power, telephone and cable come in adjacent to each other and are all surge-protected to the metal shack. There is no earth connection. Everything in the shack would be protected. The power surge protector would work fine and would not blow up. (You might not want to be entering the shack when a surge hit.)

I agree with trader.

------------------- Suppose you have a 2,000A surge current to earth and the resistance to earth is a very good 5 ohms. The "ground" at the house rises to 10,000V above 'absolute' earth potential. This happens even with very good earthing.

If the earthing is not as good the building "ground" rises to a higher potential above 'absolute' earth potential.

This can show up at equipment like a pad mounted air conditioning compressor. The compressor can be at about the potential of the earth the pad is sitting on. The power wires can be at about the potential of the building "ground". During the surge they can be thousands of volts different. A service panel protector does not prevent this. And it can happen even with good earthing.

[gfretwell]

This ground shift is a very real thing. When I was in the computer biz I have seen 35 volts between the ground in 2 buildings that were not really that far apart. In a lightning event this can be a huge number.

In extreme cases we have even run bonding conductors between buildings to protect LANs. These days an optical isolator is a better solution or just WiFi the whole thing.

I have a huge ground plane at my house (ufer, lots of rods and an in ground pool). That may be why I have survived a number of direct lightning hits. It really likes my weather station. I was in the driveway, running to the house and saw one of the hits so there was no doubt what got hit. The only thing I lost was the station hub and the serial port on the PC it was hooked to.

[HeyBub]

Right. In sum, most MOVs (Metallic Oxide Varistor) act like a fuse in reverse.

And, like a fuse, an MOV only works once (or at best a few times).

The more expensive surge protectors do not use these MOVs but instead rely on electronic magic to take care of business, then reset, ready for the next surge.

[gregz]

The devices in the protector do not last once hit. They either reduce capability or completely fail. One shot, blow up. MOV. Surge protectors or anything else, are not designed to deal with a direct lightning strike. The common connection is the same place ground is, unless it goes to a different ground stake or stakes.

Greg

[gregz]

I think that's called secondary protection.

Greg

[bud--]

That is all complete nonsense.

Not obvious what you are saying.

You do not want multiple earthing systems. You create an earthing system, which may have multiple earthing electrodes, and connect it at the service to the power system ground, which is bonded to the service neutral. Entry protectors for phone and cable must connect with short ground wires to a common connection point on the power earthing system. Dish entry protectors also connect there.

Multiple earthing electrodes can be at far different potential during a surge 'event' or a nearby lightning strike. Much of the protection is that during an 'event' the building wiring may rise far above 'absolute' earth potential, but they all rises together.

[bud--]

Nonsense.

MOVs have an energy (joule) rating. It is the energy the MOV can absorb in single event - one surge - that puts the MOV at its defined end of life (but still functional). If the energy hits are much smaller than the single event rating, the cumulative energy rating is much higher. For example a MOV might have a (single event) rating of 1,000J. If the individual hits are 14J the cumulative energy rating might be 13,000J. With high ratings a MOV may never fail.

That is particularly true for a plug-in protector. For a couple reasons the energy that can make it to the protector is very limited. An investigation found the maximum was 35 joules. That was with power service surges that were up to 10,000A (which, as in another post, is the maximum that has any reasonable probability of occurring).

Plug-in protectors with high ratings are not likely to fail. That is one reason why some of them have connected equipment warranties. If wired correctly they are very likely to protect from very near very strong lightning strikes.

(Neither plug-in or service panel protectors protect by absorbing a surge. But in the process of protecting they absorb some energy.)

According to the IEEE surge guide, "the vast majority (>90%) of both hard-wired and plug-in protectors use MOVs to perform the voltage-limiting function. In most AC protectors, they are the only significant voltage limiters." I don't know of any protection schemes for service panel protectors that do not use MOVs. They are so widespread because they have high energy dissipation capacity in a small package at relatively low cost.

[HeyBub]

Please refrain from confusing people with silliness:

"A varistor remains non-conductive as a shunt-mode device during normal operation when the voltage across it remains well below its "clamping voltage", so varistors are typically used to suppress line voltage surges. However, a varistor may not be able to successfully limit a very large surge from an event such as a lightning strike where the energy involved is many orders of magnitude greater than it can handle.

"Follow-through current as a result of a strike may generate excessive current that completely destroys the varistor. Lesser surges still degrade it, however.

"Degradation is defined by manufacturer's life-expectancy charts that relate current, time and number of transient pulses. The main parameter affecting varistor life expectancy is its energy (Joule) rating. As the energy rating increases, its life expectancy typically increases exponentially, the number of transient pulses that it can accommodate increases and the "clamping voltage" it provides during each transient decreases. The probability of catastrophic failure can be reduced by increasing the rating, either by using a single varistor of higher rating or by connecting more devices in parallel. A varistor is typically deemed to be fully degraded when its "clamping voltage" has changed by 10%. In this condition it is not visibly damaged and it remains functional (no catastrophic failure)."

What's important is that a MOV may be completely degraded without showing outward signs. Of course if the MOV is black and partially melted, you have a clue... but baring that, you just don't know.

And can't really know inasmuch as there's no way to non-destructively test the thing.

[trader4]

I think the issue here is that your statement that "devices in a protector do not last once hit" is misleading. It depends on what the rating for the protector is versus what it is hit by.

Case A: It's a 20K Amp whole house surge protector and it's hit by a typical surge that might occur from a lightning strike that occurs somewhere down the street. In that case the current/energy is going to probably be two orders of magnitude less than the rating of the protector. In that case, the surge protector survives, is still functional and it's capacity has minimal degradation.

Case B: It's a 20K Aamp protector and it's hit by a very close strike and it sees 50K amps. In that case it could either be blown up or as you say, have significantly reduced capacity.

And the far more common occurence for any protector is that it will likely see a lot more of the Case A type surges, because direct or near direct hits are rare. A direct hit can't even really reach the surge protector. If it hit the service near where it enters the house, typically a lot of the energy is going to go elsewhere, via arcing, leaving only part of the strike for the surge protector to deal with.

That does seem to be an inherent problem.