surge suppressor voltage limit

Thanks Bud I agree

BillB

Thanks again Tom

BillB

| My reference to L to L is based on an earlier comment that the 240V domestic | supply was 120VAC-Ground-120VAC ie 240V Live to Live. If this is not the | case please advise what a 240VAC domestic supply is as all my comments have | been based on this assumption.

That is the correct assumption for 240V in the USA.

| If you change the AC rated voltage of all the MOVs to 300VAC the device can | work at either 120VAC live to ground or at 240VAC live to ground. In both | cases the clamping voltage will be approximately 800V live to live (or | neutral), live to ground and neutral to ground. | Where as if the live to ground voltage is 120VAC and the neutral to ground | AC voltage is either 0 or 120V and you use a 150VAC MOV, the clamping | voltage to ground for both MOVs will be approximately 400V.

I don't see any ratings for the suppressors with regard to the MOVs other than the clamping voltage. And the value I see is 330V. Not 400V. So I assume if they are doubled, it would be 660V.

|> > If the clamping voltage in the USA is doubled, but this does not happen |> > in |> |> Europe, then the devices in both would end up being about the same, |> right? |> | |> | Yes the MOVs would be the same. |>

|> So why could the be used on 240V in Europe but not in USA? | | The answer is yes they can, but a European device will not meet US/Canadian | standards and likewise the US version would not meet European standards, as | the standards set different requirements.

Right. And that can be an issue. OTOH, the 120VAC device would only meet US standards in the 120VAC context, even if the MOV clamping voltage is doubled. So either way, I'm forced to use the device inappropriately since

240V only devices for the USA either don't exist, or do not have adequate level of protection.

|> I have some "wall warts" that are rated for 100-240V 50/60Hz, but the |> plugs |> are NEMA 1-15 (e.g. standard for 120V w/o ground pin). | | Do these have surge protection?

I don't know if they have surge protection built in. Presumably since they are rated 100-240VAC 50/60Hz, the manufacturer is saying they are good to go around the world. Could I take them to Brazil where they use the same outlet as in USA, but run 240V on it, and plug them in there? Could I plug them in using an adaptor (not a step-down transformer) in Europe? My guess is yes. That is a guess based on the printed rating. If it turns out one of these has surge protection that would trigger on the peaks of 240VAC because it is assuming 120V, despite being marked for use 100-240V, I'd be talking to my lawyer about a fraudulent advertising lawsuit.

|> |> Apparently some "experts" think that today's home appliances can |> withstand |> |> some higher surge levels, and that the MOVs are being destroyed more |> often |> |> than desirable in the protectors. |> | |> | They can, but they will last longer with the lower clamping voltage, It |> is |> | the marketing department like to use the higher energy rating as a |> marketing |> | tool, the engineer still prefer the lower clamping voltage. |>

|> But on 240V, the clamping voltage would be "just right" (doubled from |> 120V). | | What is "just right" ?

If 330V or 400V clamping is right for 120V, then 660V or 800V clamping should be right for 240V. Do you see any reason otherwise?

|> Some areas of the US have substantially more than most of the US. How |> does |> the UK compare to say, Florida? | | Its the same in the UK.

There's a part of UK with as much lightning as Florida?

|> |> 1. So maybe we don't need a lower clamping voltage in the USA, given |> that |> |> more and more appliances (especially computers) handle all of |> |> 100-240V. |> | |> | As I have already said the lower the clamping voltage the longer the |> life of |> | the protected device will be. |>

|> Of course. But I want to run the device at 240V for power efficiency. | | I have not implied that you should not use 240V, I have only tried to state | that the lower the clamping voltage is the better the equipment will be | protected.

I would agree. That makes me wonder why the "experts" recommend a higher clamping level. But in one case the implication was that the extra level of protection wasn't needed, and it was better to make the surge protector last longer over needless clamping of small surges, so it can handle the rare "big one" at least once.

|> I never do that. I always want to know who to sue (even if that would be |> a |> very unrealistic thing) :-) | | In the UK I am not as concerned over who you will sue, but about trying to | make sure you do not need to sue anyone.

And life would be easier if they just switch the USA over to 240V for all.

On Tue, 23 Sep 2008 06:26:47 -0700 (PDT) w_tom wrote: | On Sep 23, 1:38?am, snipped-for-privacy@ipal.net wrote: |>> ?Another way to reduce voltage during a surge? ?Increase varistor |>> joules. ?More joules means less voltage and less energy gets |>> absorbed. ?A concept that many have difficulty grasping. |>

|> Joules _is_ energy. ?So that doesn't make sense UNLESS you are saying that a |> higher joules RATING has a lower impedance, and that results in a lower voltage |> drop when conducting. | | Phil - you keep having this problem. Joules is energy. Does that | mean varistors work by absorbing the surge? Of course not. A

The operational aspect of the MOV is to pass the current between wires. But in so doing it _does_ absorb/dissipate energy.

| benchmark statement. More joules means less voltage and less energy | gets absorbed by the varistor. A concept that so many have diffculty | grasping, in part, because assumptions replace learning. Learn what | joules measure before posting your assumptions.

But this wording is contradictory and therefore cannot be used to render any using engineering meaning.

Joules measure energy. That is exactly what it means. So a statement that says more joules is less energy is no basis for anything meaningful.

On Tue, 23 Sep 2008 06:32:39 -0700 (PDT) w_tom wrote: | On Sep 23, 1:31 am, snipped-for-privacy@ipal.net wrote: |> Or a substantial available current sufficient to heat the MOV to the point of |> vaporization. | | If vaporization failure occurs, then a protector was grossly | undersized and was a threat to human life. Vaporization is a complete | violation of the manufacturer's "Absolute Maximum Ratings" which are | located at the very top of every MOV datasheet. Varistor must fail | only by degrading. MOVs must never get so hot as to vaporize. But | vaporization gets the naive to recommend obscenely profitable and | ineffective protectors.

I've seen the end results of a surge (I was not nearby when the surge actually took place, fortunately) that vaporized most of a (approximately) 250 kcmil cable. Are you going to tell me that any surge protector on the market that would fail to survive that is undersized?

| An effective protector remains functional after a surge so that | humans never even knows a surge existed.

There is a huge dynamic range of possible surges. The extreme surges are very rare, fortunately. But they are possible and they can vaporize things. No protection can be 100% perfect. Are you expecting 100% perfection?

| Increasing varistor voltage is to make the MOV not conduct energy | during normal operation. Varistor voltage is kept as low as possible | so that MOV does not conduct during normal operation and absorbs | minimal energy during a surge.

That's the right theory.

On Tue, 23 Sep 2008 11:40:52 -0500 bud-- wrote: | snipped-for-privacy@nospam.tesco.net wrote: |> |> I cannot see any valid reason for doubling the live to ground clamping |> voltage, it the same as saying why bother to fit surge suppression in the |> first place. The lower the clamping voltage the less likely it is to damage |> the connected equipment. | . | Francois Martzloff was the US-NIST guru on surges with many published | papers. From one of them: | "The fact of the matter is that nowadays, most electronic appliances | have an inherent immunity level of at least 600 V to 800 V, so that the | clamping voltages of 330 V widely offered by [surge suppressor] | manufacturers are really not necessary. Objective assessment of the | situation leads to the conclusion that the 330 V clamping level, | promoted by a few manufacturers, was encouraged by the promulgation of | UL Std 1449, showing that voltage as the lowest in a series of possible | clamping voltages for 120 V circuits. Thus was created the downward | auction of "lower is better" notwithstanding the objections raised by | several researchers and well-informed manufacturers. One of the | consequences of this downward auction can be premature ageing of [surge | suppressors] that are called upon to carry surge currents as the result | of relatively low transient voltages that would not put equipment in | jeopardy." | | The paper dates back to 1995. Suppressors with very high ratings are now | readily and cheaply available which may make the argument less relevant. | | For the 250V world, one would have to know the immunity level. It may be | closer to the normal voltage than in the US, making an increase in clamp | voltage less practical.

If this recommendation by Martzloff is adopted, and "immunity level" is shifted to at least 660V and maybe up to 800V, it would seem that these devices can then be used on 240VAC (340V peak). Do you know of a reason they could not be so used (other than the benefit Martzloff is expecting for 120VAC uses would not all be realized on 240VAC circuits).

| Matching would be better. At least you should sequentially go through | the paralleled MOVs, with some sharing along the way. | Or can you drive one into failure (conduction at normal voltages) before | the others are 'used up'? | | How do manufacturers get ratings of 1000+J for a single MOV in a plug-in | suppressor? Paralleling? There seem to be high rated single MOVs that | can be used in service panel suppressors.

And they are probably more expensive than manufacturers want to put into their "cost driven market" absed plug-in suppressors.

One option I am considering is to use one of those whole house protectors and put it into a box and run the power cord through it. But I don't know if this is getting me the right kind of protection for point of use.

| Another Martzloff paper looks at the energy absorption for a MOV at the | end of a branch circuit. It is surprisingly small for 2 reasons: | 1. At about 6000V (US) there is arc over from panel busses to | enclosure(+neutral+ground+earthing system). After the arc is | established, the voltage is hundreds of volts. That dumps most of the | energy coming in on hot power wires to earth. (Receptacles (US) also | arc-over at about 6kV.) | 2. The impedance of branch circuit wiring for surges greatly limits the | surge current, and thus energy, that can reach a MOV.

That impedance is quite different for common mode (higher impedance) than for differential mode (lower impedance). In differential mode, there is an equal and opposite current channel in parallel, confining the magnetic field, and limiting the inductance. For mixed mode surges, the common mode will be impeded greatly, leaving differential mode.

I notice from videos of the 33kV surges that happened to homes in Harford county MD, severe damage levels was propogated into at least the inside breaker panels. If the arcing at the meter and the service cable does not completely quench the surge, and it gets to the panel and destroys that, then why not also go beyond and take out appliances? The question is just how much voltage reached each point. Arcs have a voltage drop, but they do also have impedance (inductance). So they cannot completely clear a differential surge. Clearly the arc at the meter did not clear the surge since there was more surge damage at the panel.

| The maximum energy dissipated in the MOV was 35 Joules for a 10 meter | branch circuit. In 13 of 15 cases it was 1 Joule or less. That was with | surges source currents from 2,000 to 10,000A (the maximum likely for a | home). Surprisingly, the highest energies were for some of the lower | surge currents because the MOV could hold the service panel voltage | below arc-over.

Is this maximum based on the value at the END of the circuit, or at the entrance to the home? I do know surges MUCH greater than 6000V can hit a home.

Purpose of wire is pass the current between locations. But in so doing, wire _does_ absorb/dissipate energy. Both wires and MOVs do the same thing better when less energy is absorbed. Superior is when wires and MOVs divert/transfer/conduct more energy and absorb less energy.

How to make a protector absorb less energy during a surge? Increase its joules. More joules means less voltage and less energy gets absorbed by the varistor. A concept that so many have difficulty grasping, in part, because assumptions replace learning.

Again, the V-I charts are in the datasheets. These above concepts are obvious once numbers from V-I charts are grasped.

Many foolishly feel that a protector works by absorbing surges. They *know* this because protectors are rated in joules. If true, then more joules in a protector means the protector absorbs more joules. They *knew* by replacing facts and numbers with assumptions. More joules means less energy gets absorbed by a protector.

A protector that absorbs less energy is the superior protector. A protector's job is to divert / connect / shunt / bond /conduct that energy into earth where surges get absorbed / dissipated harmlessly. Not to block or absorb energy. The less energy a protector absorbs means even more energy gets absorbed / dissipated harmlessly in earth.

But again, many have been so enthralled by what is promoted in retail stores as to *know* a protector works by absorbing surges. Many, using assumption, have difficulty grasping what MOVs do. Forget what was promoted by retail salesmen. Learn from this V-I charts in datasheets. The better protector with more joules absorbs even less energy.

You would have to add protection for any signal wires that run to the protected equipment (not difficult). .

. A MOV at the end of a branch circuit is differential mode.

If there was a common mode surge on the power, at the end of a branch circuit the return would be the power system ground, which is an adjacent conductor. .

. It was not a surge.

It is difficult-to-impossible to design practical protection for a power line cross.

Thanks for separating out .mpgs from this event [another thread]. (But I haven't downloaded them yet.) .

. "With surges *source* currents...." Most lightning produced surges act as current sources. The source current went to a panel which fed a branch circuit. One point of the explanation above is that there is arc-over at panel at about 6kV. The branch circuit sees a pulse over

6kV, the arc starts, the source voltage drops to hundreds of volts. (Except for some smaller surges where the MOV was able to keep the voltage at the panel below 6kV as I said above.)

The 'experiment' was actually run on simulation software which has been validated with physical experiments.

Phil

I give up you don't seem to understand any of the comments I make I hope someone else has more patience then me. We could go on for ever and you will not understand the topic. Like said I give up on the topic.

BillB

Leeds Lad in exile

| I give up you don't seem to understand any of the comments I make I hope | someone else has more patience then me. We could go on for ever and you will | not understand the topic. | Like said I give up on the topic.

Maybe this is because you were never really trying to answer the exact question I asked, and instead were trying to answer something else. Maybe that something else is something you thought was more important. And maybe it is. But you did not connect it to the topic. I am only focusing on the issue I raised and am intentionally ignoring other issues, unless it can be shown they have specific bearing on the issue of interest.

Maybe if you narrow down the points you think I misunderstand instead of trying to clump a bunch all together (that always makes things more confusing to follow through with).

I've dealt with a few people here who seem to have a lack of understanding (especially those guys that just quote other people's stuff in circumstances where it doesn't apply, hoping to win points with facts). But you do seem to be someone that understands stuff. I think the problem might be you are not expressing yourself well enough in terms of relating it to the context at issue.

On Wed, 24 Sep 2008 08:07:20 -0700 (PDT) w_tom wrote: | On Sep 24, 2:17 am, snipped-for-privacy@ipal.net wrote: |> The operational aspect of the MOV is to pass the current between wires. |> But in so doing it _does_ absorb/dissipate energy. | | Purpose of wire is pass the current between locations. But in so | doing, wire _does_ absorb/dissipate energy. Both wires and MOVs do | the same thing better when less energy is absorbed. Superior is when | wires and MOVs divert/transfer/conduct more energy and absorb less | energy.

Wires and MOVs pass current, not energy. Maybe this is the point of confusion.

If I am passing 5 amps over a wire, tell me how much energy that is in a exact numeric value of joules.

| How to make a protector absorb less energy during a surge? Increase | its joules. More joules means less voltage and less energy gets | absorbed by the varistor. A concept that so many have difficulty | grasping, in part, because assumptions replace learning.

Joules is energy. You want to INCREASE it to accomplish DECREASING it?

I think you are misunderstanding the difference between Joules and Amps.

| Many foolishly feel that a protector works by absorbing surges.

That actually is ONE way to make protectors. It, by itself, is not very effective because it generally is a one-time use. It is not a practical method for home and most business usage.

| They *know* this because protectors are rated in joules. If true, | then more joules in a protector means the protector absorbs more | joules. They *knew* by replacing facts and numbers with assumptions. | More joules means less energy gets absorbed by a protector.

The ratings are understood to mean the surge energy level that can be protected against. The term "absorb" might be misused (probably is). But this says nothing about what will be absorbed and/or diverted by any given protector.

| A protector that absorbs less energy is the superior protector. A | protector's job is to divert / connect / shunt / bond /conduct that | energy into earth where surges get absorbed / dissipated harmlessly. | Not to block or absorb energy. The less energy a protector absorbs | means even more energy gets absorbed / dissipated harmlessly in earth.

No.

Earth is not the only target for diversion.

A protector can suppress a differential surge by simply providing a path for it to reach its opposing voltage. No earth is involved. Making such a path operate at as low an impedance as possible makes such protection both more effective and more survivable (to live on to protect again).

A protector can protect against a common mode surge by spreading the surge across a zone of interconnected equipment. Low frequency common mode surges are more common (because the inductance of the path up to this point of use impedes the higher frequency components more), so equalization can be effective to keep damaging current levels from taking place on equipment interconnections.

| But again, many have been so enthralled by what is promoted in | retail stores as to *know* a protector works by absorbing surges.

I'm not going to add the topic of "consumer misunderstanding of what terms like absorb means, or consumer misunderstanding of how surge protectors do their job". It's quite clear to me that even _you_ do not understand all the things involved.

| Many, using assumption, have difficulty grasping what MOVs do. Forget | what was promoted by retail salesmen. Learn from this V-I charts in | datasheets. The better protector with more joules absorbs even less | energy.

What does "with more joules" mean to you? You seem to be misusing the term.

On Wed, 24 Sep 2008 10:08:43 -0500 bud-- wrote: | snipped-for-privacy@ipal.net wrote: |> On Tue, 23 Sep 2008 11:40:52 -0500 bud-- wrote: |> |> | Matching would be better. At least you should sequentially go through |> | the paralleled MOVs, with some sharing along the way. |> | Or can you drive one into failure (conduction at normal voltages) before |> | the others are 'used up'? |> | |> | How do manufacturers get ratings of 1000+J for a single MOV in a plug-in |> | suppressor? Paralleling? There seem to be high rated single MOVs that |> | can be used in service panel suppressors. |> |> One option I am considering is to use one of those whole house protectors |> and put it into a box and run the power cord through it. But I don't know |> if this is getting me the right kind of protection for point of use. | . | You would have to add protection for any signal wires that run to the | protected equipment (not difficult).

Of course. Right now I am doing that by means of wireless isolation. So there are no metallic signal wires involved. Everything will be in one zone, powered through one power supply path with the protection in place. If I end up needing signaling at higher data rates than wireless can do, then I will use optical.

|> | Another Martzloff paper looks at the energy absorption for a MOV at the |> | end of a branch circuit. It is surprisingly small for 2 reasons: |> | 1. At about 6000V (US) there is arc over from panel busses to |> | enclosure(+neutral+ground+earthing system). After the arc is |> | established, the voltage is hundreds of volts. That dumps most of the |> | energy coming in on hot power wires to earth. (Receptacles (US) also |> | arc-over at about 6kV.) |> | 2. The impedance of branch circuit wiring for surges greatly limits the |> | surge current, and thus energy, that can reach a MOV. |> |> That impedance is quite different for common mode (higher impedance) than |> for differential mode (lower impedance). In differential mode, there is |> an equal and opposite current channel in parallel, confining the magnetic |> field, and limiting the inductance. For mixed mode surges, the common mode |> will be impeded greatly, leaving differential mode. | . | A MOV at the end of a branch circuit is differential mode. | | If there was a common mode surge on the power, at the end of a branch | circuit the return would be the power system ground, which is an | adjacent conductor.

Unless the common mode is across ALL wires including the EGC. That is where w_tom thinks it has to go to earth. But in fact it can be reflected back up the circuit.

A surge that is arriving on one wire and not the other wire(s) is a mixed mode surge.

|> I notice from videos of the 33kV surges that happened to homes in Harford |> county MD, severe damage levels was propogated into at least the inside |> breaker panels. If the arcing at the meter and the service cable does not |> completely quench the surge, and it gets to the panel and destroys that, |> then why not also go beyond and take out appliances? The question is just |> how much voltage reached each point. Arcs have a voltage drop, but they |> do also have impedance (inductance). So they cannot completely clear a |> differential surge. Clearly the arc at the meter did not clear the surge |> since there was more surge damage at the panel. | . | It was not a surge.

So do you classify anything happening to utility equipment as NOT a surge?

| It is difficult-to-impossible to design practical protection for a power | line cross. | | Thanks for separating out .mpgs from this event [another thread]. (But I | haven't downloaded them yet.)

And yet these are things that do happen a lot, although not generally as spectacular as what happened in Harford county that one day. I remember reading about one such event when I was around 12 years old. Based on what I remember from that reading, it sounded like all conductors into the author's home were energized at the same polarity, or else the voltage was not as high. More often the crosses that happen are 7200 to 8000 volt lines dropping down onto lower wires, the 120/240 network, or even phone or cable wires. Then those make give the MV a new path to ground.

|> | The maximum energy dissipated in the MOV was 35 Joules for a 10 meter |> | branch circuit. In 13 of 15 cases it was 1 Joule or less. That was with |> | surges source currents from 2,000 to 10,000A (the maximum likely for a |> | home). Surprisingly, the highest energies were for some of the lower |> | surge currents because the MOV could hold the service panel voltage |> | below arc-over. |> |> Is this maximum based on the value at the END of the circuit, or at the |> entrance to the home? I do know surges MUCH greater than 6000V can hit |> a home. | . | "With surges *source* currents...." Most lightning produced surges act | as current sources. The source current went to a panel which fed a | branch circuit. One point of the explanation above is that there is | arc-over at panel at about 6kV. The branch circuit sees a pulse over | 6kV, the arc starts, the source voltage drops to hundreds of volts. | (Except for some smaller surges where the MOV was able to keep the | voltage at the panel below 6kV as I said above.)

I am also looking for ways I can protect against the very strong surges, from a direct hit of lightning onto the meter panel, to a "Harford count event". The latter case would be the most difficult (as you point out) because of the large current and continuing supply. A simple MOV would start conducting, but the supply from a 33kV line would very quickly vaporize that thing. An arc tube would likely protect more, but even it would be soon overcome.

But that is not the topic of THIS thread. It would make an interesting thread on its own. Maybe I'll ask that (again) some day (or you can, if you want).

One thing I do wonder about in the Harford event is how much internal arcing was happing in the transformers intended to step the 13/8kV down to 120/240 because of 33kV being applied. If you have a transformer designed to step

13.2kV (dual bushing) down to 120/240, what is the rated BIL? How high a voltage is it tested with?

A surge is, by definition, a very short event - certainly well under a cycle. The next most damaging surges, after lightning, are generally from utility switching. Crossed power lines like Hanford are far too long duration and are "temporary overvoltage" or maybe "swells" (up to a few seconds).

MOVs are rapidly destroyed by long duration events. Martzloff has written "in fact, the major cause of [surge suppressor] failures is a temporary overvoltage, rather than an unusually large surge."

On Thu, 25 Sep 2008 09:35:39 -0500 bud-- wrote: | snipped-for-privacy@ipal.net wrote: |> |> |> I notice from videos of the 33kV surges that happened to homes in Harford |> |> county MD, severe damage levels was propogated into at least the inside |> |> breaker panels. If the arcing at the meter and the service cable does not |> |> completely quench the surge, and it gets to the panel and destroys that, |> |> then why not also go beyond and take out appliances? The question is just |> |> how much voltage reached each point. Arcs have a voltage drop, but they |> |> do also have impedance (inductance). So they cannot completely clear a |> |> differential surge. Clearly the arc at the meter did not clear the surge |> |> since there was more surge damage at the panel. |> | . |> | It was not a surge. |> |> So do you classify anything happening to utility equipment as NOT a surge? | | A surge is, by definition, a very short event - certainly well under a | cycle. The next most damaging surges, after lightning, are generally | from utility switching. Crossed power lines like Hanford are far too | long duration and are "temporary overvoltage" or maybe "swells" (up to a | few seconds).

My definition of a surge is bigger than your definition of a surge!

| MOVs are rapidly destroyed by long duration events. Martzloff has | written "in fact, the major cause of [surge suppressor] failures is a | temporary overvoltage, rather than an unusually large surge."

Then they need to design for that.

My definition of surge is industry standard including the IEEE. It works better if you use the same definitions other people do. .

. As has been covered in this newsgroup before, it is not practical to protect from temporary overvoltage with MOVs. Only practical protection I know of is to disconnect from the source. Depending on how high the overvoltage that may not work well.

On Fri, 26 Sep 2008 09:16:26 -0500 bud-- wrote: | snipped-for-privacy@ipal.net wrote: |> On Thu, 25 Sep 2008 09:35:39 -0500 bud-- wrote: |> | snipped-for-privacy@ipal.net wrote: |> |> |> |> |> I notice from videos of the 33kV surges that happened to homes in Harford |> |> |> county MD, severe damage levels was propogated into at least the inside |> |> |> breaker panels. If the arcing at the meter and the service cable does not |> |> |> completely quench the surge, and it gets to the panel and destroys that, |> |> |> then why not also go beyond and take out appliances? The question is just |> |> |> how much voltage reached each point. Arcs have a voltage drop, but they |> |> |> do also have impedance (inductance). So they cannot completely clear a |> |> |> differential surge. Clearly the arc at the meter did not clear the surge |> |> |> since there was more surge damage at the panel. |> |> | . |> |> | It was not a surge. |> |> |> |> So do you classify anything happening to utility equipment as NOT a surge? |> | |> | A surge is, by definition, a very short event - certainly well under a |> | cycle. The next most damaging surges, after lightning, are generally |> | from utility switching. Crossed power lines like Hanford are far too |> | long duration and are "temporary overvoltage" or maybe "swells" (up to a |> | few seconds). |> |> My definition of a surge is bigger than your definition of a surge! | . | My definition of surge is industry standard including the IEEE. It works | better if you use the same definitions other people do.

My interest is in protecting against that which I have termed "surge" under my definition. If you have a better term to use that covers my definition, please suggest it. Until then, the only term I have is the one I have been using, which basically covers any kind of overvoltage situation other than utilities setting the voltage too high. And this is the term most CONSUMERS would use over this broad range of definition, and the scope of protection they expect.

|> | MOVs are rapidly destroyed by long duration events. Martzloff has |> | written "in fact, the major cause of [surge suppressor] failures is a |> | temporary overvoltage, rather than an unusually large surge." |> |> Then they need to design for that. | . | As has been covered in this newsgroup before, it is not practical to | protect from temporary overvoltage with MOVs. Only practical protection | I know of is to disconnect from the source. Depending on how high the | overvoltage that may not work well.

It's also silliness to expect any ONE SINGLE COMPONENT to provide the entire scope of protection. For example, I don't expect an MOV to protect against extremely fast rise times. Low pass filters are for that. Protection needs to come in the form of a system that provides all different kinds of protection within reason (the incident in Harford county _may_ be a case that is not practical to protect against at a level most consumers want to deal with). The system needs to work together so that what is done to provide one form of protection does not end up degrading other forms of protection. For example, we must have the grounding electrodes correctly wired at the point of entrance and not connected at other points in the building (with certain exceptions that won't generally exist for residences).

. I did. .

. If you want to use defined terms in nonstandard ways maybe you should post on a non-engineering newsgroup. It is not possible to communicate when there is no common language (surge) or science (phil's phantasy physics). . > |> | MOVs are rapidly destroyed by long duration events. Martzloff has

. MOVs are faster than surges.

On Sat, 27 Sep 2008 10:00:37 -0500 bud-- wrote: | snipped-for-privacy@ipal.net wrote: |> On Fri, 26 Sep 2008 09:16:26 -0500 bud-- wrote: |> | snipped-for-privacy@ipal.net wrote: |> |> On Thu, 25 Sep 2008 09:35:39 -0500 bud-- wrote: |> |> | snipped-for-privacy@ipal.net wrote: |> |> |> |> |> |> |> I notice from videos of the 33kV surges that happened to homes in Harford |> |> |> |> county MD, severe damage levels was propogated into at least the inside |> |> |> |> breaker panels. If the arcing at the meter and the service cable does not |> |> |> |> completely quench the surge, and it gets to the panel and destroys that, |> |> |> |> then why not also go beyond and take out appliances? The question is just |> |> |> |> how much voltage reached each point. Arcs have a voltage drop, but they |> |> |> |> do also have impedance (inductance). So they cannot completely clear a |> |> |> |> differential surge. Clearly the arc at the meter did not clear the surge |> |> |> |> since there was more surge damage at the panel. |> |> |> | . |> |> |> | It was not a surge. |> |> |> |> |> |> So do you classify anything happening to utility equipment as NOT a surge? |> |> | |> |> | A surge is, by definition, a very short event - certainly well under a |> |> | cycle. The next most damaging surges, after lightning, are generally |> |> | from utility switching. Crossed power lines like Hanford are far too |> |> | long duration and are "temporary overvoltage" or maybe "swells" (up to a |> |> | few seconds). |> |> |> |> My definition of a surge is bigger than your definition of a surge! |> | . |> | My definition of surge is industry standard including the IEEE. It works |> | better if you use the same definitions other people do. |> |> My interest is in protecting against that which I have termed "surge" under |> my definition. If you have a better term to use that covers my definition, |> please suggest it. | . | I did. | . |> Until then, the only term I have is the one I have been |> using, which basically covers any kind of overvoltage situation other than |> utilities setting the voltage too high. And this is the term most CONSUMERS |> would use over this broad range of definition, and the scope of protection |> they expect. | . | If you want to use defined terms in nonstandard ways maybe you should | post on a non-engineering newsgroup. It is not possible to communicate | when there is no common language (surge) or science (phil's phantasy | physics).

If you can't understand what is being communicated, then don't respond. I've communicated by definition of surges before. Any time you need to be reminded, just ask.

| > |> | MOVs are rapidly destroyed by long duration events. Martzloff has |> |> | written "in fact, the major cause of [surge suppressor] failures is a |> |> | temporary overvoltage, rather than an unusually large surge." |> |> |> |> Then they need to design for that. |> | . |> | As has been covered in this newsgroup before, it is not practical to |> | protect from temporary overvoltage with MOVs. Only practical protection |> | I know of is to disconnect from the source. Depending on how high the |> | overvoltage that may not work well. |> |> It's also silliness to expect any ONE SINGLE COMPONENT to provide the entire |> scope of protection. For example, I don't expect an MOV to protect against |> extremely fast rise times. | . | MOVs are faster than surges.

So what. That doesn't make them perfect. There are two issues:

The leading edge of a fast rising surge wavefront will NOT propogate only to the MOVs in the hopes that the MOV will provide a low impedance path somewhere. The true fact is that the wavefront will propogate in all directions in some proportion dictated by the reactive aspects of the wiring involved.

The common mode surge will present no voltage differential across the MOV at the point of use.

A very fast rising common mode surge, which would typically result from a potential induced directly into branch circuit wiring, would bypass MOVs and go directly to appliances. Differences in distance (timing) and reactance in the paths to different appliances in the point of use zone protected by a given protector can result in brief high voltage differences between these interconnected appliances. Additionally, branch wiring within the appliance can result in differences within.

Of course bud will just pretend these not-so-common types of surges do not ever happen.

Dear Phil

While I have no intension of being drawn into discussions relating to this topic, I would like to bring to your attention an excellent publication by the IEEE that covers most aspects of surge suppression technology as it relates the protection of your home. Please see "How to Protect Your House and its Contents from Lighting", this publication covers both the requirements under US standards and why some systems work better than others.

BillB

Leeds Lad in exile