Grounding of Lightning Protection Systems

Codes and standard practice is to provide individual earth pits for lightning, electrix and PLC, the earth pits would be bonded together later for an equipotential bonding.

What your distributer says is incorrect, the lightning flash is of very high intensity / frequency, it does not dissipates immediately on ground contact, it would follow the path of least resistance and over a matter of time situation normalises.

Regardless of anybodies opinions here, you should follow your local code as specified by your local inspector.

I've seen both separate and common grounding pits for lightning protection and building structure. What seems to be more important is that regardless of where the building ground pit resides, the lightning protection down lead should go immediately into a grounding pit upon reaching ground level. Most lightning protection systems though are not designed to take a direct strike. They're designed to reduce the likely hood of a local direct strike. The few systems (radio antennas mounted on a mountain ridge line) that are designed to handle a direct strike are very expensive systems. All enquiries on this black art must go to the engineers, geologists and meteorologists that designed that system.

I've been away from the field for awhile but your distributor was correct. The reasoning was as follows:

The lightning arrester system is designed to allow the discharge the "bolt" before it occurs. The air terminals are pointed to facilitate the discharge before a bolt occurs. Sometimes this doesn't happen and the bolt occurs. As was pointed out, the bolt may not completely egress through the lightning discharge system, and enter the building through conduit, downspouts etc. That is why the systems are connected. In addition, sometimes the bolt will hit the nearby service pole and enter through the power, telephone cables, etc.

Motorola used to have a standard called R56 that dealt with this issue.

While the "code" is nice and gives you CYA , it doesn't necessarily mean that you have a best practices scenario for protecting life and property.

Well, I'll offer this from the National Lightning Safety Institute...

"High impedance paths must be avoided since they can create unwanted heat and/or mechanical damage. Adjacent metal bodies, for example, must be bonded to the low impedance path "so as to prevent side flash or spark-over" (see 780, section B.2.2).

Where down conductors are near metal bodies without bonding the two, lightning tends to jump from one to the other tending to create sparks (see section B.3.9). For more information, see Annex B of NFPA-780."

http://www.lightningsafety.com/nlsi_lhm/grounding_principles_NFPA780.html

The NEC requires that the down conductors be bonded to the grounding system. I suspect you did that when you bonded them to the building steel. That is all I would do providing that the steel is part of the grounding system.

" It also has leads coming up to bond to an MCC, main breaker and an automatic transfer switch"

I believe that is already bonded to the system. No need to do it again.

If the building is of steel it will act like a "Faraday cage" in which case do we need lightning protection system for it. The steel columns will act as down conductors. If the roofing sheets are metal will there be any problem,will it attract lightning?.

Regarding earth pits, for each service they have specified a different value of earth electrode resistance. Will some one quote values from codes/standards(NFPA/BS/IEC)?.For example for lightning protection it is 10 ohms but if building contain explosives it should be 6 ohms. For electrical power the resistance depends on the earth leakage or fault relay setting provided the touch voltage do not exceed 50 volts. Similarly for telecom and computers thre should be recommended values. Also they have "clean earth" for some applications.

pnaban, I have issues,

The steel columns will act as down conductors. The steel columns will act as down conductors.

I have a problem with this. If the bolt follows this path it will end at the bolts mounting the columns to the concrete. It will not travel any further and who knows what happens?

I've seen copper plates in the ground destroyed by lightning strikes and the only way to find it was electronic equipment operating erratically. When tested, Ground plates all failed every test using an Oscilloscope.

I have personally witnessed black lightning once in my life. It was amazing and changed my persona of lightning altogether after discussing what I seen with a meteorologist friend of mine. I could only describe it as black lightning coming up from the ground to meet the sprite from above. And when it did, major fire works.

In steel buildings the four columns at the corners are usually earthed as well as few more at the centres of the length of the building depending on the dimensions of the building. The question in this thread was "whether down conductors are required for steel building".

No pnaban, the question is whether individual grounding systems should be bonded as recommended by the LP consultant.

The answer is yes they should be. The bond should be capable of disassociation to permit testing of individual systems.

In the UK, we install a pit to each down conductor and also for the building. These are then interconnected through removable links at the main earth terminal, thus allowing testing of the systems individually. If you have sensitive electrics within the structure then I would also recommend a Surge Protection Device at the incoming mains and communication systems. However even as some have pointed out this would not stop your building being struck by lightning, it is just an easier path for the lightning to follow.

Regardless of electrical/electronic grounding system, the lightning grounding system must be separated. All downlead grounding conductors from lightning rod arrester shall be directly to a grounding rod in the ground. The grounding rod / pit shall be 2.5m minimum distance away from the building structures. Nevertheless refer to a local applicable codes in your place.

Although each system(power/telecom/computer/lightning etc) has its own earth pit/rod ,they are interconnected automatically by a device(name forgotten) when a lightning strikes to reduce the combined earth resistance.

Don't you love language open to interpretation? This means of course that every inspector has his own opinion as to how this is to be implemented.

The term "bonded" simply means connected. But as a matter of practical application, a grounding system for lightning should offer plenty of surface area in which the energy can be dissipated before it is connected to the electrical system.

Otherwise, every air duct, conduit, and chassis of everything plugged into the electrical system effectively becomes a lightning rod.

My case in point, a number of years ago, I was standing barefoot on a return air duct holding my infant daughter, at the glass walk out door watching the rain when lightning struck across the street. The strike was nearly 100 feet away but we both received a nasty jolt through the vent. So now comes the argument that we were either struck as well, or that the electrical ground became part of the dissipative circuit. The shock was about the level of a 220 VAC accidental contact. I really shouldn't know what that feels like but I do.

Even if the lightning struck the common neutral line somewhere (by the way the electrical lines in that area are underground), there should have been sufficient grounding along the way so that we were not affected.

So, if it were up to me, I would still have a separate ground rod driven into the ground for the electrical power distribution system and then "bond" it with the lightning rod system ground. Hopefully it is significant (as in several connected rods close together or a large mesh).

Lastly, install surge protection. Whether the ground is raised above or below the power line voltage, the difference will be minimized. I would look for a system capable of shorting up to 20,000 amps for a few microseconds without failure. They are quite common.

Mike:

Appreciate the answer and all of the answers that were given. The fact is, the NEC does require bonding the lightning protection system grounding to the electrical service grounding. Some have suggested that a separate electronic ground be created, but the NEC prohibits that in Article 250. Some local inspectors will defer to good engineering judgment if you can make your case, but here, I think I have designed a very low impedance electrical ground (probably will test less than 2 ohms resistivity) and that should suffice. I also will make the lightning protection ground bond connection outside the building to a 1/0 copper bare in the footer trench at 2 points. Also am providing a Category C 125kA surge suppressor on the 480v mains at the MCC. That should be enough to protect from gross impulses.

Again, thanks to everyone for writing in.

Phil in Kentucky

Mrxphill:

Seems you've opened one of the more interesting discussions. There has been some anecdotal points, as well as various interpretations of the NEC code. Don't come down too hard on the code writers, they are engineers who've done extensive studies on lightning including after action reports. One major company did tremendous research in Miami (one of the lightning capitals) in the 80's. It even looks as if many contributors have a consensus, really unusual. The problem is that even with all the study starting with Ben Franklin, we have little idea of what a particular bolt is going to do when it gets out of bed. I doubt there is any 100% solution, nor do I think there is a 100% perfect installation. Therefore, get your lightning installer and the various equipment reps together and determine what is best. The lightning installer will know the local codes. If you follow the code, your insurance company will be happy, OSHA will be happy, and the building will pass inspection.

Oh and ask the installer the following. Lightning has a very fast rise time. This means that a large flat conductor is the best way to route a bolt to earth. Then why do they use braided wire as a downlead? I've never gotten a good answer to this.

The last question on braided wire I can answer. Braided wire is used not because of an electrical quality but because of a mechanical quality. The repeated flexing and movement that an exterior cable will be exposed to from thermal expansion, wind, natural mechanical resonance, ice expansion and other forces will mean that this cable be allowed to move or it will break. A broken cable will not prevent ion build up (the actual task of lightning protection). Unfortunately a broken cable can promote a direct lightning strike by bringing an almost isolated conductor into a thunder storm.

Don't be tempted to deploy these types of air terminations.

They reduce to a shower of sparks in the event of a strike. This renders them inoperative and can start secondary fires.

The system must be able to survive a high energy shunting event and continue functioning after the fact.

First of all, what is the purpose for earthing ?. Earthing is all about protecting and preventing the user of electricity from getting electrical shock.So for me i will suggest, you can go ahead to earthing if only the user can reach the metal case.

The simple thumbrule is

1. Use separate ground for grounding Electronic Equipment

2. Use separate ground for Electrical Equipment grouding & together with System Grounding.

3. Use separate groundind for Lightning.

None of the abobe to be connected with one another. The GPR for lightning strike may be quite affecting if connected to Electrical Equipment grounding and may cause damage to equipments etc.

For steel structure if the steel sheet thicknes is above 4.86mm no separate air terminal is necessary (Ref NFPA 780).

Some do connect Lightning & Electrical Grounding together, but its better not to do so.

I you don't bond the grounding systems then step voltages will occur that will guarantee a toast session in the event of a strike (near or direct). Not as you have stated.

The down conductors (drains) should be segregated for each functional earth but all bonded at or below ground level.

Where bonding of grounding systems may introduce noise into noise sensitive apparatus then the bonding is still performed with transient earth clamps which are open circuit under normal operating conditions but go short circuit when a voltage spike is presented to it by a discharge induced step voltage. That way you get isolation normally and bonding when required.

Bond or burn. You choose.

Can you write back which standard considers lightning strike Grounding Potential rise in calculating step potential?

AS 1768:2007 - Australian Standard for Lightning Protection

In the UK, it is BS 62305 covers this topic.