Impedance under lightning current

[Leyden]

i'm trying to research what might be the difference in two electrical scenarios

scenario
same in both scenarios-
we have a 18"tall x 4"deep x 8" wide carbon steel box, this is mounted to a building and electrically bonded to the electrical system throughout the building and most of the rest of the conductive material in the building. in the middle of the box there is a 6awg solid copper wire bonded to the box and

scenario 1 variation-
the wire exits through a 1/4" diameter opening without touching the box where it exits

scenario 2 variation-
the wire exits the box in a connector that bonds the wire to the box where it exits

same in both scenarios-
the wire runs down to a grounding electrode in contact with earth

background on question-
-when a conductor is surrounded by ferrous material it creates a choke which impedes current flow on the conductor
-the solution used for this problem is bonding the conductor where it exits its surrounding ferrous material whether once or twice. this creates a parallel electrical path, using the enclosure as the dominant current path in this case
would anyone be able to help me with what the impedance might be on the conductor in both scenarios under lightning current (say around 100 million volts, 200KA, 10MHz)
obviously there are many variables that i have not given a value to, i am only looking for ballpark values and what you all think the differences may be

more background info on the question-
-the purpose of the wire is to connect the electrical system to Earth and stabilize the voltage to Earth when the system undergoes a surge
-the root of the question is, the less voltage gradients (even if only marginally less) throughout the system the less damage will be done to things, also the less time (even marginally and almost immeasurably less) the less damage and negative effects will be done by the lightning strike. so will eliminating the little choke help?

any good info, direction or thoughts are greatly appreciated, even if your not sure, although if your not sure please say so
thanks

 

[Baluncore]

The question will be how the lightning gets into the box. It is far more likely that it will pass over the outside of the box to reach the building earthing structure.

The 100 MV will never appear on the building structure. That is the cloud to Earth voltage prior to the strike.

The current risetime will be limited by conductor inductance. High inductance will make for greater voltages in the lightning conductors. V = L* di/dt. You refer to a 10MHz BW, say a faster than 50nsec risetime to 1kA. That suggests transmission line analysis will be needed and skin effect will be important. Use bare flat conductors which do not corrode and have low inductance.

A strike will follow the surface of a conductor unless there is a sharp bend, when it will cut across the bend through the air.

Any conductor with insulation is a liability with lightning. The velocity factor will slow the internal transient sufficiently to allow an external current to flow, which will destroy the insulation.

Bonding an Earth conductor where it passes through a metal box wall is a liability. It appears counter-intuitive, but skin effect guarantees the inner and outer surfaces of the box wall are not connected together at RF. The strike will go elsewhere.

Using an insulated grommet or gland is a problem because when the strike occurs, it will flash boil any moisture in the gap and then burn or eject the seal, or cut the through wire.

You will need to provide spark gaps outside the box so lightning has somewhere to go, other than into the box.

Designing systems to minimise lightning damage starts out as magic, then a black art. It becomes a science once you have seen enough damage, analysed the mistakes and revised your ideas.

 

[Leyden]

thanks for the great info

Bonding an Earth conductor where it passes through a metal box wall is a liability. It appears counter-intuitive, but skin effect guarantees the inner and outer surfaces of the box wall are not connected together at RF. The strike will go elsewhere.

this connector actually bonds on the exterior of the enclosure and to the exterior portion of the conductor. wouldn't it be better to have the grounding electrode conductor bonded to the exterior of the enclosure (like doing so with this connector)?

just for some more background now that i see you are extremely familiar with this, this question is being asked predominately because of NEC 250.64(E) as to whether it is acceptable to run the GEC through the 1/4" hole already made by manufacturers in meter bases for the use.

i understand there are far better lightning protection schemes, but I'm trying to determine whether running the GEC just through the 1/4" hole with nothing is better or worse than using a bonding fitting that bonds the exterior of the enclosure to the exterior portion of the GEC, the GEC runs to the grounding electrode system. this is about the bare minimum code requirements.

thanks again for the great information, would you be able to recommend any publications that i may be able to study to better my understanding?

 

[Baluncore]

With a 1/4” hole you do not stand a chance of handling a real lightning strike. I think you should only expect to conduct a fault current, or guide the collected electrostatic charge to ground. You might consider cutting a bigger hole with an insulated gap so a fault current can get in and out of the box. If the same hole held all phase conductors and the protective ground, the sum of all currents through that hole would be zero.

You might use a galvanised box. NEC 250 is not used here, but I would expect the GEC to be bonded to the box at entry and at exit. That would permit a lightning strike to run over the exterior of the box, which it will probably do anyway. Keep lightning conductors separate from the power distribution.

Re:commended reading, is difficult to specify as you must build on your experience. I would suggest you start by googling 'lightning' etc.

Without some understanding of skin effect and electromagnetic surface currents it will be difficult to predict what will happen. You will need to change your view of electrical conductors, from “pipes carrying electrons”, to “mirror surfaces guiding electric and magnetic fields”. The energy travels in the insulation between the wires that guide the poynting vector, which is important to understanding energy flow.
https://en.wikipedia.org/wiki/Poynting_vector#Interpretation

 

[sparkie]

I just wanted to chime in a bit here from a practical perspective. I am also assuming typical installations (for instance, typically a painted steel box is used rather than an unpainted stainless steel box). There is one key thing that is being left out here. You are bonded within the enclosure with a bonding jumper and some type of lug assembly. The service entrance grounding conductor is also #6, which can be solid OR stranded, the difference being the amount of surface area contact. The more surface area you have in contact with the equipment grounding conductor, the less resistance you have, which is the path that the current will want to take.

That is the primary reason you do not necessarily want a fitting on the 1/4" KO that is stamped into the enclosure. Not only is it unnecessary, but you don't want multiple points on your enclosure with a solid ground. You want to provide a path with as little resistance as possible to Earth as soon as possible so that you don't have crazy things happen, like a lightning strike going through the power system of your home rather than straight to ground. You also do not want the enclosure acting any more like a conductor than it already is. This is another reason we scrape paint from around bonding jumper contact points.

 

[Leyden]

I just wanted to chime in a bit here from a practical perspective. I am also assuming typical installations (for instance, typically a painted steel box is used rather than an unpainted stainless steel box). There is one key thing that is being left out here. You are bonded within the enclosure with a bonding jumper and some type of lug assembly.

second sentence of my first post says its bonded in the enclosure, i don't believe you are offering a practical perspective but the common misunderstanding.

The service entrance grounding conductor is also #6, which can be solid OR stranded, the difference being the amount of surface area contact. The more surface area you have in contact with the equipment grounding conductor, the less resistance you have, which is the path that the current will want to take.

That is the primary reason you do not necessarily want a fitting on the 1/4" KO that is stamped into the enclosure. Not only is it unnecessary, but you don't want multiple points on your enclosure with a solid ground. You want to provide a path with as little resistance as possible to Earth as soon as possible so that you don't have crazy things happen, like a lightning strike going through the power system of your home rather than straight to ground. You also do not want the enclosure acting any more like a conductor than it already is. This is another reason we scrape paint from around bonding jumper contact points.

i recommend you read what Baluncore posted in this thread. the fitting would be electrically bonded to the exterior of the enclosure. you have to take skin effect into consideration. bonding the GEC to other things it is associated with is not a problem, if you run the GEC through a ferrous enclosure it must be bonded on both ends to use the enclosure as a parallel electrical path (high surge currents will barely use the conductor in the ferrous enlcosure).NEC 250.64(E)

 

[sparkie]

250.64(E) is referring to when one is using a race. Far more often than not, conduit is not used on the jumper from the grounding electrode to the primary service ground. You mentioned running a #6 ground wire through a 1/4" KO in the enclosure, indicating that there is no raceway. As far as the conduit bonding goes, the AHJ has a lot of say in how the bonding is done. For instance, in my old town they accepted locknuts as a bond, however here they require a grounding bushing to be used and bonded at both ends of the race in order to ensure electrical continuity.

 

[Leyden]

250.64(E) is referring to when one is using a race. Far more often than not, conduit is not used on the jumper from the grounding electrode to the primary service ground. You mentioned running a #6 ground wire through a 1/4" KO in the enclosure, indicating that there is no raceway. As far as the conduit bonding goes, the AHJ has a lot of say in how the bonding is done. For instance, in my old town they accepted locknuts as a bond, however here they require a grounding bushing to be used and bonded at both ends of the race in order to ensure electrical continuity.

250.64(E) Raceways and Enclosures for Grounding Electrode Conductors
250.64(E)(1) ferrous metal raceways and enclosures for grounding electrode conductors shall be electrically continuous from the point of attachment to cabinets or equipment to the grounding electrode and shall be securely fastened to the ground clamp or fitting. ferrous metal raceways and enclosures shall be bonded at each end of the raceway or enclosure to the grounding electrode or grounding electrode conductor to create an electrically parallel path.