I live and work in the southwest, where we experience serious seasonal monsoon storms with a lot of lightning. As a result, we have to deal with power surges quite a bit, not only from the storms, but with the POCO switching around grids during the subsequent repairs.

Due to this, TVSS's are a growth market in the area, primarily for computer protection in all office situations. In studying the suppliers tech data on the various units available, I've found a hole in the available information. Perhaps one of you could fill me in;

1. What is the efficiency comparison between quality type plug-in units compared to a panel mount?

2. Which is better? A higher, or lower "clamping Voltage" rating.

3. From the data that I have read, (supplied by the manufacture's ), is it really adviseable to combine panel mount TVSS with Plug-in strip type tVSS?

Well I am no expert on this but living in florida the lightning capital.We use surge arrestors at the panel and surge suppressors downstream.I took a hit last month with a arrestor on my panel guess it went bad but it wiped out my phones my low voltage lighting my secruity system tripped my 200 amp main but it did'nt get my cpu.I use a true sine wave ups that I plug into a suppresors and then a plug it into a gfci.The gfci seems too go bad but that is it.Anyway the way I see it the more the merrier.It is also nice to get one with a indicator.I don't think they can handle too many spikes.

Not sure how you would measure "efficiency" here, but a unit at the service entrance is FAR superior to ANYTHING installed at the point of use.

A TVSS relies on a short, heavy, low-impedance path to earth ground. The only reasonable place to get this is at the main service panel. A plug-in unit relies on the ground path back through the branch circuit wiring, which presents too much impedance to be of any value with fast-risetime events like a lightning strike.



Lower, as long as the clamping voltage is higher than the peak voltage seen under any normal condition. In an area subject to high line voltage, a unit with a very low clamping voltage may be destroyed by firing continuously on mormal line voltage.


It certainly wont hurt anything, but the point of use units are pretty useless by themselves. They do often contain RFI filtering components, which may help in an interference situation.

This is an area where, if ignorance is bliss, I am feeling quite happy right now.....

But it seems to me that the two serve separate functions, and would compliment each other.
The TVSS at the panel would protect you from stuff coming from the PoCo side of things, while the point-of-use stuff will protect you from stuff created within the house....by, say, a loose connection on a shared neutral.

surfinsparky:

Very true. Most units have MOV's (Metal Oxide Varistors) as the primary surge components, they are deliberately designed as "sacrificial" elements. Open up a strip that has failed and look for the remains of a disc-shaped component, that's your animal.
Their ability to absorb surges weakens with repeated surges. For a critical application in a highly-surge prone area, annual replacement would be a good idea. High-quality units have fuses which will open when the arresting capability is exceeded, protecting the connected equipment and the strip from catastrophic failure.

renosteinke:
Probably, maybe, sort of. I say this from experiences, because a lot rides on the clamping voltage rating of the protector.

Case #1 was a P.A. amplifier on a portable genny. Standard surge strip, after a few hours of run time, the generator suddenly labored and almost stalled, this accompanied by smoke from the surge strip. Killed genny, opened strip to see the Line to neutral MOV burned out. Restarted genny, found that due to overspeed, it was outputting 158 volts!! P.A. amp was quite hot but not damaged.

Case #2 was an open neutral situation. Surge strip burned up, damaging the floor it was on, and the connected equipment (Sat dish, VCR, DVD and Surround Receiver) were all destroyed. After the fact (who knows how high voltage really got) measured 194 volts with my Fluke 36.

Many surge strips will still let through harmfully high voltages to the equipment with little or no clamping action. Perhaps one of the higher-grade with the fuses I mentioned above would've done better, but there is still a time factor which makes me think that they still wouldn't protect adequately.

American Power Conversion "APC" makes many high-quality products, including some line voltage regulators and UPS' which automatically bring out-of-range voltage to safe levels. Extreme overvoltage results in a shutdown to protect the equipment.
http://www.apc.com

In Case #2 I used their regulator to help protect against any further problems.

In any event, as others have stated, it doesn't hurt to combine panel-mount and point-of-use TVSS anytime there could be surge issues.

Thank you all for your clear, concise, and very helpful answers.

It never fails to amaze me that even after 20+ years in the trade, there is always something new to learn every day.

I can tell that this forum will be a valuable tool to enhance my business in the future, and I hope to be able to contribute in return as well.

Thanks again,

Matt

Nobody else has said it, but the lead length between the TVSS and your panel buss bar is critical. Most recommend 6" or less...this is harder than it seems to arrange in your typical residential panel.

The best setup is a unit that goes between your meter and your socket. However, most EUSERC utilities won't allow this. Second best is a unit that sits directly on the bus bars of your panel. Depending what you are dealing with, they make ones that look like breakers and pop on, all the way up to factory built in units that bolt next to the main lugs. Third best is an external unit that sits next to your panel. I have used ones from Leviton that have plug in surge modules, allowing for quick and easy replacement. They are also fused to prevent a n extended short. The way I usually hook these up is to order a meter main with a 200A breaker, and a 50A breaker, and run the 50A to the TVSS, and the 200A to a subpanel.

NJWIRENUT said:
"A TVSS relies on a short, heavy, low-impedance path to earth ground".

Sorry, but earth or ground has nothing to do with TVSS and has no function. Any TVSS device is just a clamp to fire at a prescribed voltage and acts as a short circuit between the two points it is connected too.

For example lets say you install a TVSS at the service entrance on a single phase 240 volt service. The best units (I am talking thousands of dollars) only have two surge protection (SPD) devices installed. One each from from line-to-neutral. You will not find a device installed L-G or N-G. Absolutely no need for one.

Once You get down stream from the N-G bond at the service, a SPD between L-G and N-G becomes usefull, but earth or ground still has no function. All you are doing is clamping the voltage between L-G or N-G at the receptacle or point of use. The EGC impedance is an open circuit to earth at lightning frequencies.



[This message has been edited by dereckbc (edited 12-13-2005).]

If you are looking for some easy to read and understand info about TVSSs I suggest you to take a look at the following site: http://www.surgex.com/library/42001.html

For external transients suppression, panel mounted are the best option (Class “C” TVSS for ANSI/IEEE C62.41). There are many manufacturers available. I’ve used Liebert products in several occasions and they performed good (up to now). If pace of mind is important, you should provide both normal mode and common mode protection (LL, LN, LG, NG).

If you are concerned about budget (who’s not_?) you could (but I’m not recommending it) sacrifice LL protection because high external transients are more likely to be lighting induced common mode surges (all lines coming into the facility measured against ground). If the PoCo or another external facilities introduce in your panel high differential mode (surges between lines) transients, the LN, LG and NG sections of your TVSS together with a good grounding system will handle the job. If lightning is a concern you should check NFPA 780.

Hope this help,

Joe.-

Joe of NJ, if you are referring to a service entrance TVSS on a grounded system (even ungrounded), common mode is a waist of material, time, money, and serves no purpose.

The common mode is already provided in the "normal mode" via the MBJ on grounded systems. You cannot get any more protection than a dead short piece of copper. Only downstream of the entrance should common modes be used. As stated earlier ground (earth) has no function or purpose with TVSS devices.

Hello “dereckbc”:

You are confusing the “common mode” term with the “normal mode” one. “Normal mode” suppression is “somehow” present in the “common mode”, can you see why? Draw the diagram. But I’m sure for most commercial TVSSs, this “somehow” I’ve mentioned set us outside the IEEE 587 guidelines.

Second, the only point in which NG protection is irrelevant is very close to the N-G bonding point, but only as long as that bonding remains intact. Depending on the needs it could be necessary to have NG protection even for Class C TVSSs (most integrated units I have seen include it).

United Power has a simple yet interesting article about this topic: http://www.unitedpowercorp.com/technical/ModularSurge.htm.

I’m wondering myself if being so “definitive” as you were could drive someone for the wrong way. Maybe we should be more careful with our comments. It could be better to point the direction instead of dictating the solution.

Regards,

Joe.-

Joe of NJ, don't mean to step on your toes, just trying to de-myth manufactures claims. I know very well the difference between common and normal modes, and at a service entrance where you have a MBJ there is no common mode, or better said comon mode and normal are the same...

I have drawn it out and designed TVSS units for GE. Here is a link of some white papers I published that relates directly to common mode and normal mode.
http://www.geindustrial.com/products/applications/service_entry.pdf

Here is a quote:

There are no real benefits of choosing a 7-mode TVSS (asuming 3-phase service) over a 4-mode TVSS at service entry when the N & G are bonded except for allowing a manufacture to quote "per phase" maximum surge current ratings.

The rest of the publications can be found at http://www.geindustrial.com/cwc/pro...p;id=tvssoem&typeId=2&lang=en_US

Let’s see, common mode transients are those occurring between lines and ground (LG and NG). Yes, neutral is a line, but if we don’t want to consider it a “line” then we have to differentiate two kinds of common mode transients, the ones between lines and ground (LG) and the ones occurring between neutral and ground (NG).

Normal mode transients are those occurring between each line and the others (LL and LN). Again, if we don’t want to consider the neutral as a line we should recognize two kinds of normal mode transients, the ones between lines, and the ones between lines an neutral.

At the system bonding jumper, neutral and ground became one, so in both cases (considering the neutral a line, or not doing so) the transients could appear between lines and this neutral-ground entity (LN = LG, common mode transients) or between lines (LL, normal mode transients). I don’t see why you insist in the non-existence of common mode transients.

By the way, normal mode transients may be called transverse mode transients or differential mode transients.

The paper you wrote, shows the difference between 7 and 4 devices TVSSs, both of them in common mode configuration. In your paper you show why when neutral and ground are bonded the use of a 7 device TVSS is equivalent to the use of a 4 device one. That could be theoretically correct, but that has nothing to do with normal mode versus common mode (or, if it does I don’t see it).

We have to remember that according to NEC 250.30 A 1, the system bonding jumper “ …shall be made at any single point on the separately derived system from the source to the first system disconnecting means or overcurrent device, ….”. So, there is no assurance that at the Class C TVSS the neutral-to-ground path is 0 ohms (or something equivalent) at the transient spectrum and considering the surge current capabilities. 300 amps only require an impedance of 1 ohm to produce 300 volts! I could agree with you in that for Class C TVSSs installed at service entrance the NG TVSS device could not be of much help in many applications; but I disagree with the idea of it being totally unnecessary in all kind of situations.

But all this is not too much critical at least we are thinking on individual TVSS devices (individual, 2 leads elements), because even those GE offers in the website you pointed at shown NG protection. I know, that protection is given by the series of TVSS devices between LN-LG, but anyway they are promoting it, so I’m afraid you need to keep working in your de-myth project, to make it really transparent (and not only concentrate in de-mystifying other manufacturers products while hiding your own myths [by the way, GE is not the only one doing this, most manufacturers may be doing the same]). So, if it is not possible to evade this NG protection, why bother arguing it is not necessary and it is a waste of time and money?

Please, this is not personal, I don’t want polemics, but as I stated before, we should be more careful with that solution-designing feeling that want to control us from time to time.

Joe.-

Yins guys are sounding pretty high brow so I might as well jump into the fray. One thing I think I've learned is that no matter how many times I tell lightning how it's supposed to behave, it never seems to listen. Ponder my thoughts and feel free to blast away at me. I won't feel hurt.

Lighting strike = single turn primary??

Area between lightning and house = air core??

House wiring roughly parallel to strike = loosely coupled, single turn secondary??

Main panel surge suppressor = sitting there whistling, doesn't know that the furnace ctl. board just blew up, again. Doesn't know that it's going to get blamed for not stopping a surge that didn't come from the utility.

OK, blast away, make it pithy & no bloviating. ( I'm watching waaay too much O'Reilly.)
Joe

Joe of NJ, please consider these items.

1. According IEEE 95% of all surges happen on the primary side of a transformer in the common mode. If that is the case then the surge appears on the secondary as a differental or transvers mode (L-L & L-N

2. At the service entrance the neutral and ground are bonded together. There is no SPD that can equal that short circuit. If you were to add a SPD between L-G or N-G would only parallel the L-N device. Therefore adding devices L-G and N-G anywhere there is a N-G would be waisted material and ineffective. You get more bang for the buck by using a larger SPD device than adding multiples in parallel.

3. When you bond the N-G you essentially make the differential mode and common mode one in the same becuase they are both the same point electrically at the N-G bond. I never implied that transverse and common were the same, only at the service entrance. Please go back to my first post where I sated all-modes were usefull downstream of the N-G bonds, about 80-feet to be exact.

4. Common modes does become important downstream from the entrance and downstream from SDS for the Class B and A units (sometimes called Point of Use). But earth in no way has anything to do with the operation of a TVSS which was my original point. Joe test, that should address your concerns as of the induced strike.

X

[This message has been edited by XtheEdgeX (edited 01-16-2006).]

XtheEdgeX


No that is not what they are installed to do.

What you have is a basic misconception here. Voltage is not 'dispersed' to ground. It must be directed back to the source, the earth is not the source of the transient voltages.

Voltage does not 'want to' go to ground it 'wants' to go back to the source.

The source will be (most times) the first transformer on the supply side of the TVSS.


[This message has been edited by iwire (edited 12-31-2005).]

X

[This message has been edited by XtheEdgeX (edited 01-16-2006).]

Sorry I still think you are mistaken as to what the earth will do for a TVSS.

Please re-read Dereck's posts, he is a very knowledgeable man.

Bob

X

[This message has been edited by XtheEdgeX (edited 01-16-2006).]

X

[This message has been edited by XtheEdgeX (edited 01-16-2006).]

Here is a newly published document that contains a wealth of information and is not promoting any particular mfg or product. Effective surge protection requires a systematic approach that involves ALL electrical and communications systems.
http://www.panamax.com/PDF/IEEE_Guide.pdf

Strange, that is my thought as well.

XtheEdgeX,

The quote you attribute to Mr. Tajali is being used slightly out of context. His comment was made during the discussion of solidly grounded, resistance grounded and ungrounded distribution systems. His point is that low-resistance wye connections make TVSS applications/installations easier, not that a low resistance to ground is required to make a TVSS work.

From the other references you supplied, I get the impression that all of the authors are using the colloquial term "ground" rather than the more correct terms like "grounding conductor" or "reference point".

As a PE I have no problem with the information provided by Dereck.

X

[This message has been edited by XtheEdgeX (edited 01-16-2006).]

Xedge chill out and open your mind for a moment than you are free to respond.

No one has said those PEs you listed are wrong.

What I believe is you are misunderstanding what they are telling us.

For the most part when these PEs say 'grounding is critical' they are not talking about the connection to earth but the interconnection of the EGCs.

The only time a connection to earth may help a TVSS is if it is dealing with a lightning strike.

All other transients must go back to the source which is not the earth.

You are correct there are MOVs from line to ground.

You are mistaken as to which way those are directing current.

Those line to ground MOVs direct transients on the ground to the line and back to the source.

If you tried to direct a utility supplied transient to 'earth' all it would do is find its way back to the source through other grounding electrodes. It is more efficient and of less impedance to direct it right back to line.

Anyway, that has been my understanding of TVSS operation and I have installed a few myself.

A simple question.

Would you say a TVSS installed in an airplane would not work as there is no connection to earth?

Bob



[This message has been edited by iwire (edited 01-03-2006).]

X

[This message has been edited by XtheEdgeX (edited 01-16-2006).]

XtheEdgeX, since you will not post anymore, maybe you might listen for minute.

Functionally there is no difference between Gas Tube, MOVs, and SAD's. They are do the exact same thing, they are all shunt devices. Meaning they appear as an open circuit (or load) until a threshold voltage is reached, at which point they become low impedance devices (or short circuit). In other words they are all shunt devices dsigned to be installed in parallel with the load they are protecting.

The only differences between the three is the amount of energy they can dissapate measured in Joules (Gas having the highest, and SAD's the lowest), the response time in which they operate measured in fraction of a second (SAD's being the fastst, and Gas being the slowest), and robustness or durability (Gas is the most rugged, while MOV's & SAD's are prone to failure from operating). Other than that they operate in the same manner, a shunt.

As I stated earlier in a gounded service N and G are the exact point at the disconnect, they are bonded together. Is that clear? No device type (Gas, MOV, or SAD) can match the performance of a bolted fault. Therefore adding any type of device from N-G or L-G at the entrance can add anything of value to the user. It is a waist of material and money. Only modes needed at a grounded service entrance are L-L and L-N, Ground is already provided by the N-G bond.

Once you are downstream from the entrance then L-G and N-G modes become useful. However earth still has nothing to do with the circuit. The impedance of the EGC and the series impedance of the electrode appear as an open circuit to the fast rise time of a transient. What any SPD does do is clamp or limits the voltage between the two points it is connected too to a safe or reasonable limit. So in the case of a terminal strip TVSS you would select the SPD clamp voltage for the N-G mode to be something like 10-volts, and L-N and L-G to 330-volts.

Dereck Campbell, PE

X

[This message has been edited by XtheEdgeX (edited 01-16-2006).]

Do they make surge protectors for people? No.
Calm down, this is a discussion area. If you want to trade insults, do it with e-mails.
Alan.

X

[This message has been edited by XtheEdgeX (edited 01-16-2006).]

X transients arriving via the service is a differential mode rather than common mode, meaning the voltage is L-L and L-N. Earth impedance is of no value or importance, it is merly a reference point.

"X" as a moderator at ECN I must ask you to edit your post or have it edited for you.

Profanity is not tolerated here at all.

Spirited discussion and disagreement sure.

This web site is strictly rated "G".

Bob

There is a certain amount of truth to that, but turn it around.

Because we all know Dereck we have a fair idea of the depth of his knowledge.

We don't know anything about XtheEdgeX and frankly, right or wrong that name makes me think of skate boarder.

So far what sticks with me is your quickness to fly off the handle.

Bob

Dereck: X transients arriving via the service is a differential mode rather than common mode, meaning the voltage is L-L and L-N. Earth impedance is of no value or importance, it is merly a reference point.

So you can then look at it as the equipment sees it: a connection to 3 points that are labelled,"L", "N", & "G". We know that the other ends of "N" & "G" were at the same potential at a main panel and that the equipment ends may or may not be. But if we look into the equipment for any components between N & G, there aren't usually many susceptable ones there. You might find a few HF bypass or decoupling caps or maybe a spark gap or two. But in the over 25 years that I have repaired a broad range of equipment, I have rarely had to replace components L-G or N-G. My forgetter is getting as good as my rememberer so it's possible I'm forgetting a few HV discs I replaced in the 80's and just remember the half dozen or so input bridges and switching mosfets I've replaced in the past year.

If I'm given a hypothetical choice to use 2 components to protect my gizwalter, I'll pick a bi-directional transorb and a fast acting fuse. The closer the clamp voltage above the highest L-N expected, the better. So once in a blue moon, your mains drift up and you pop your fuse. So what! My fuse is not going to be much higher than my max I and definately picked to blow before the surge and continuous ratings of the transorb are exceeded. If I'm allowed a 3rd component, I would pick series inductance in the line after the transorb before I would clamp L-G or N-G. I think any good electron lawyer would claim that surges delayed are surges denied. Scott posted several drawings of surge arrestors that were just pi-type LC input filters with MOVs in parallel with the caps.

I'm curious if other folks out there who have to fix things after they blow have had similar experiences with the components that they have had to replace.
Joe

Regarding the TVSS discussion taking place:

As mentioned previously, the intents of many TVSS devices installed on the Load side of the Main Disconnect, are to "Divert" Surge(s) from the "Active Circuit", with hopes that the devices connected to the normal Active Circuit (L-N) will not have to deal with any surges.

The idea is to attempt to pass surges, which originate from sources both external and internal to the system, around the sensitive load item(s), and back to the Power Supply.

Except for Lightning surges and RFI (or even line charging...but that's stretching it!), the surges will be between Source and Load - or Transformer and Equipment.

This would mean that any TVSS component connected to the "Grounding Conductor", outside of the Point Of Physical System Earth Bonding, is simply shunting the surge load around the normal active circuit, back to the Point Of Physical System Earth Bonding - and eventually returning to the Power Supply of origin.

BTW, Point Of Physical System Earth Bonding is in reference to the location on a given AC Power System, where the System's Grounded Conductor is Bonded to the local Grounding Electrode System - via the Grounding Electrode Conductor(s); along with the Metallic Enclosures and Equipment Grounding Conductors being Bonded to the same "Star Point".

Surge Protection Devices (SPDs) on the "Line" side of the Point Of Physical System Earth Bonding, will still - in effect - be connected the same as others - unless they are not connected to the same Electrode as the AC System - and this Earthed side is isolated from any physical connection to the Bonded AC System's "Grounding System" (GES and Enclosures).

In our field, the term "Grounding" and "Grounded" are very misleading - and at times downright confusing; to the point of pending disaster!

I encourage using terms like "Bonded", "Bonding" and "Equipment Ground Bonding", in the field, so the idea of a physical connection to the AC system is made - instead of the notion of several mysterious connections of wires to the Earth.

There is (typically) only one connection of the AC system to the "Earth Ground", and this is only to establish a "Neutral Point" - not for any system performance reasons.
The system will work exactly the same with no physical Earth connection - and this includes TVSS devices (as long as the EGCs still terminate to a given Conductor of the referenced AC system).

I need to create an easy to follow seminar of basic system grounding principles - how, what, where, when and why scenarios to describe the ideas of AC System Grounding, and submit this to the Technical Reference section.

This is really important, and once again, I have had to resolve some major confusion of Grounding Techniques in the Field - between Clients, Contractors and our own Employees.


Per a part of "Joe Testengineer's" last message


This is 100% correct! And goes along with the first part of my reply.

The TVSS designs I submitted on-line (posted in the Technical Reference section), are - by the most part, simple PI filters - with MOVs in Parallel (between L-G and N-G).

The intent of these TVSS devices are to shunt surges away from the Load devices, by "Dumping" the surges into the Equipment Grounding Conductor, and "Sending Them" back to the Source / Supply via the connection of the system's Grounded Conductor to the Bonded Equipment / GES Star point.

On Ungrounded AC Systems, the termination of TVSS devices, along with Low Pass / Band Pass filtering terminating to an Equipment Grounding Conductor, will still eventually result in flows returning to the Power Supply (Transformer).
They will be limited to the level of flow which would be Capacitively Coupled to the System.
Or in the cases where the Secondary side of the PoCo's Transformer is Earth Grounded via a Center Tap point only (and there is no system Grounded Conductor), there will be a flow through the Earth Ground, between the local Grounding Electrode System(s) and the PoCo's Transformer Earthing Conductor point of connection to the actual Ground.
I will include a few drawings and examples + explanations of this type of Ungrounded AC System.

OK, time to leave the soapbox!

Scott35

Joe, I agree with your assessment. To summarize L-G, N-G modes are not necessary at the service entrance, and earth ground has no real function in the operation of a TVSS device. Earth is just a reference point and is used for safety. Down stream from the entrance N-G and L-G modes become useful because a lot of sensitive electronic equipment has components like caps, mov’s, and RFI filters installed in these modes for FCC requirements. Again, the earth ground has no function. However, the EGC does become a factor b/c if we were to have a surge current flowing in the neutral circuit and/or EGC, this would cause an excessive voltage drop between N-G, therefore a SPD between the 2 would limit the potential difference to a safe operating limit felt across the equipment terminals

Here is a passage of what I have written in the Emerald Book:
“Low-voltage end-user type surge protective devices are often described as transient suppressors, but their operation is really a diversion of the surge current through a low impedance path preventing the rise of high voltages across the load terminals. For large surge currents, this diversion is best accomplished in two stages. The first diversion should be performed at the entrance to the building, typically by conventional surge arrestors rated for this duty (a class “C” device). Then, any residual voltage resulting from the action of the arrestor can be dealt with by a second protective device at the power panel, or at the terminals of a connected load (a class “B” or “A” device) . Note that class “A” devices are sometimes called point of use devices.”

SCOTT, I looked at your drawings and correct me if I am wrong. Your devices appear to be class “A” devices aka Point of Use devices. I agree the LC network would dampen any transient on the load side, and recognize the value of the N-G and L-G modes. These would great devices on the service side except you would not need any of the N-G and L-G modes. But the problem is they would have to be service rated equipment and very expensive to pass service current ratings of 100 + amps. In theory I know they are possible, but I have not run across any in my adventures. In fact your devices look a lot like the ones used in Triplett Surge Reference Equalizers for Point Of Use devices. A very fine product indeed.




[This message has been edited by dereckbc (edited 01-18-2006).]

OK, I obviously started something here, and I will try and make ammends. With TVSS, I was taught that in order for it to function properly, it had to have a good low impedance path to ground to deal with any excess voltage that the clamping device didn't take care of.
I'm sorry for anyone who may have gotten a bad impression of me. Sometimes I get a little stubborn about my work related stuff. If you were to be debating something like this with me in person, and I love to debate, you would see that I will convey my opinions this way, but actually not be upset, as it may have seemed. Kind of like ribbing a co-worker. For example, if I was talking about this with one of the guys at work, he may say to me "Shut up, you don't know what you're talking about". And I might say "Who do you think you are? You're just an apprentice". LOL. Does that make sense? But, when I sat back and read my posts, I saw that some things could have been offensive. I apologize for that. It's hard to convey feelings like that when you write it down. Anyway, I'll try and get on the right foot, and choose my words more carefully. I would also like to add, that if I have been taught a way that someone doesn't think is correct, I won't change my thoughts just because someone tells me I'm wrong. I would appreciate it if I could be directed to something more convincing. If I say this is how to do this, and you say no it's not, look at NEC section such n such, I will see that I was wrong. Or direct me to a IEEE standard or an OSHA reg or something. I've encountered too many people that would show someone the wrong way, that's all.

X, no problem, lets move on. Think about this for a while. Let’s go back to your assumption that a low impedance earth ground is a factor, before I do that, some good reference material is IEEE STD 1100-1992 (aka The Emerald Book POWER GROUNDING SENSITIVE ELECTRONIC EQUIPMENT). I am co-authoring the 2006 release and have some insight. Also NFPA -780 is excellent material.

Let’s assume we go to the trouble of installing a ring ground with chemical rods and get that elusive 5-ohm or less ground. Now we bond our service with a lot of overkill using a 10-fott long piece of 750 MCM ground electrode conductor. Most people would think this is a killer system no lightning could damage. Would you agree?

Now let’s apply some basic physics and electrical principals. Lightning and surges are high frequency events so inductance and impedance comes in to play. So let’s assume the event is lightning and it hits the weather head, the worst case scenario. Due to the very fast current rise time of lightning the impedance of that 10-foot long 750 KCM cable becomes a few kilo-ohms, for our example let’s say 2000 ohms. 2000 ohms is in series with our killer 5 ohm ground electrode system would be 2000 + 5 = 2005 ohms at the N-G bond point. Do you see that logic?

If so, I now pose the question to you. What difference does it make if the ground electrode system impedance is 5-ohms or say a two rod 50-ohm system? If it were a 50-ohm system the N-G bond point would be 2000 + 50 = 2050. Is there any significant difference between 2005 and 2050 ohms? Now for the reality bites section; The 5 or 50-ohm ground electrode measurement is measured at DC or power frequencies which have nothing to do with High Frequency (HF). At HF the impedance would be many magnitudes higher depending on the frequency of interest.

Now for a final thought or two. If lightning were to hit the weather head as we used in our example, the voltage at the N-G bond would be several thousand volts above local earth ground. Let’s just say 10,000 for argument sake. Although the reference ground point (the N-G bond) goes to 10,000 volts, so does our Line and neutral voltage. The job of the TVSS is to clamp the voltages between terminal points to a survivable level. Since N-G is bonded by a bolted connection that mode and L-G mode is covered at the service entrance only. All we have to do is clamp is the L-L and L-N modes that will be felt by the terminal equipment. If we clamp the L-L to say 400 volts, and L-N to 330 volts we should be OK. Yes the voltage would rise to 10,000 on all circuit conductors with respect to local earth ground but not between the EGC-N-L that the equipment is seeing.

Do not going away thinking a low impedance ground is not important, because it can be when we are talking about lightning protection systems. We will save that topic for another time.

Hope that helps.

Dereck

X glad to have you back, I did not mean that you had to pull all your posts, only the words you would not use with children.




A stubborn electrician.....what a shocker.

I certainly am guilty of that.

Not many of us here are not the same way, it seems to go hand in hand with pride in the job we do.

Anyway the past is history, today is a new day.

Bob

X,

Also glad to have you back. I was looking forward to more of the "healthy" debate/discussion on this topic.
I mean like this post ain't any where near a record length yet.

dereckbc,
OK, your last post made more sense to me. Especially: this "Although the reference ground point (the N-G bond) goes to 10,000 volts, so does our Line and neutral voltage". My thoughts were that the excess line surge needed to be re-directed, and to ground, but if the ground has the same surge current on it...Also, I'd like to add that I'm in Florida, and you know we have lot's of lightning here. But, a recent study conducted in Tampa (I'm not sure of the date of this study), shows that 80% of the surges came from equipment inside the premises, while only 20% came from lighning and the utlility. So this led to my other belief, that a surge from inside, had to be directed to the outside of the premises. And when I'm talking about a surge re-direction, I'm talking about the excess voltage that escaped the clamping devices. If I remember correctly, the article said that only 20% of the surges were from outside sources, because of how effective modern day service entrance TVSS is.

X, glad I was able to help out. FWIW you are now on to something; most transients are generated within, it is true in commercial and industrial applications. That is why the entrance TVSS is just a starting point called Class “C” application. You have to apply Class “B” and Class “A” devices to control the internally generated transients.

Again I will say earth ground has no function here either. But that is not to say the EGC does not have a function, because it does. Internally generated transients are all carried on the premise wiring. Any transients need to be returned back to its source. So how do you do that. To start you install Class “B” devices in all breaker and distribution panels. Actually the best method is to install panels with TVSS built-in the unit. Add-ons can be made worthless by the installation itself. Finally you install Class “A” device right across the point of use or at the terminals to stop (or return it) at the source.

Some more IEEE/ANSI document that you might try to get your hands on is IEEE Std C62.41 Recommended Practice on Surge Voltages in Low-Voltage AC Power Circuits, IEEE Std C62-1990 Guide and Standards for Surge Protection, and UL-1449.

This has been an outstanding thread!!!
It is good to see XEdgeX come back and explain himself...good for you... it will be good to hear your ideas in the future.
Bob- it is good to see you refraining from putting the hammer on X. Who says you are stubborn?

Derec... as usual, you are great. Each time I learn more and more from you and my peers will benefit as well.