Lightning Strikes and design theory 1

[guilio2010]

So I wanted to get some feedback or ideas on a recent lightning storm that caused damage on some equipment.

During a lightning storm, a din rail of containing terminal blocks was damaged. In Pic 2 of the din rail, you see where a shield terminal block was connected but the contacts burnt through the rail (Pic1). On one end of the shielded terminal block was connected to an isolated ground bar. The ground wire that was connected to this was connected to a isolated ground bar. That wire was replaced due to it being damaged (reports mention it was broken in half). This ground bar has a 2AWG leading from the cabinet (Pic4) and travels under ground to off location to a ground grid, but during the run, it taps into and connects to some of the equipment ground. The next photo (pic2) shows one side of the din rail that is blacken, that feeds into the controller cards. Behind the wires (pic3), you see a screw where the din rail arced through. Pic5 shows a bolt that connected to the subpanel and some blackness around it. The door was blown open. Majority of the instruments connected to the RTU was replaced due to damage. No other din rails has signs of damage as extent as this.

Going to the basics here is my thought process.
-For the ground wire to be blown in half, the voltage had to exceed the maximum specs. This would be equivalent to inducing a high pressure on a pipe. The pipe will bust.
-The current being induced on the din rail melted the contacts from the shield terminal blocks to the din rail.
-The current induced on the din rail began to melt with the screw and once separation occurred with the screw to din rail, arcing began due to voltage levels.
-The wires going to the controller on the lefts side were blacken, and I think this is due to the arc flashing. However, there were ground wires connected to a card in the RTU, but the wires are only blacken on one end.

Here are the questions I have:

1) Where could the source originate at? I think there was a very close strike, where the lightning struck the ground, went through the ground wire and into the cabinet before being dissipated through the ground again. I have arguments stated that lightning will not come through ground since it can not travel up. I get the philosophy as if you take a high pressure pipe filled with product and open a valve to atmosphere, the liquid can not travel back into the pipe once into atmosphere, but again, the lightning is a source of power, so why can't lightning send power in the ground, energize the ground wire and loop through? Others feel it was a device or equipment, but no signs exist of a direct lightning strike.

2) I understand the philosophy lightning goes to ground and dissipates, but I argue the point if lighting strikes the ground directly, it does energize that ground and if the resistance of 2 paths exist, then the current will be proportional to that path of resistance. Hence if you take lightning strike near a ground wire, then the ground wire will pick up the power, but since the ground wire is large, the resistance is very low compared to other sources, it mostly travels back to ground. Thus, the other end may have some blow fuses if designed correctly.

3) Why would only 1 din rail see this type of extensive damage but the other don't? I would expect the same type of burning effect on the other din rails. Same with the isolated ground, only 1 wire burnt up.

Thanks,
guilio

 

[Power0020]

Interesting!

The breakdown of insulation and air can take place once the voltage exceed a certain level, it is a probability thing that we can't be 100% sure it will happen or not.But a figure named (basic insulation level) BIL gives a 10% probability, and another figure (critical flash over)CFO gives a 50% probability of (breakdown) BD.

The DIN rail had a weakest insulation point to ground and flashed at that point, discharging the lightning induced voltage/current to ground.

The point here is that you have isolated ground, which isn't "real" ground for DC system (DC usually run ungrounded). The voltage between the isolated ground and real ground exceeded the BIL level of the air-gap between the din rail and the housing of the panel and flashed at that point (maybe this was the point with smallest gap?).

The origin of lightning may be somewhere in the vicinity of the panels with cables running into the panel getting the voltage but not breaking down. The dilemma of over-insulation is that it will transfer the surge to another weak point and flash there (that is why we need insulation coordination).

It may also have struck the building and discharged to ground but in its way, it induced voltages on nearby cables terminating at the panel.

To get a grasp of that, see Dehn lightning protection guide available on this link :

http://www.dehn-international.com/s...f/blitzplaner/bpl2015/lpg_2015_e_complete.pdf

 

[1capybara]

You asked:
1) Where could the source originate at?
high speed cameras show many light strikes are upward from the earth to the clouds.

2) I understand the philosophy lightning goes to ground and dissipates, but I argue the point if lighting strikes the ground directly, it does energize that ground and if the resistance of 2 paths exist, then the current will be proportional to that path of resistance. Hence if you take lightning strike near a ground wire, then the ground wire will pick up the power, but since the ground wire is large, the resistance is very low compared to other sources, it mostly travels back to ground. Thus, the other end may have some blow fuses if designed correctly.
lightning causing a voltage rise in the ground plane is well known behavior

3) Why would only 1 din rail see this type of extensive damage but the other don't? I would expect the same type of burning effect on the other din rails. Same with the isolated ground, only 1 wire burnt up.
Lightning is unpredicable, we cant say what it will do, only what it is likely to do. Like Power0020 said! Think about this - instances of lightning jumping from the lightning downconductor (connecting "lightning rods on the roof the ground rods or ground ring) to a metal rain downspout, then back to the downconductor.

 

[guilio2010]

Thanks for you time all. I know lightning strikes are very hard to distinguish, but from my end, I get a lot of force that people want to jump to this conclusion because it's the easiest to understand, and not really look into it to see if that is what really happened.

Power0020:

Thanks for the link. It's going to take me a little time to go through it, but I'll try and get what information I can and see if I can relate it to our situation. From your post:

What do you mean insulation coordination?

For the BIL from the din rail to cabinet, I agree as evidence shows this, but I guess I'm wondering about the shield terminal block also. In the previous picture set (Pic1), the contact points of the terminal block where it comes in contact with the din rail are gone. If current exceed this, I understand. I guess I'm trying to figure out if the surge came from the iso ground bar, through the shield terminal block wire, through the terminal block to the din rail, then arc on the real ground screw attached to the cabinet. I guess I'm looking at also if arc situations, does the metal from high potential eat away to the low similar to galvanic cathodic protection where the anode eats away to protect the cathode? I guess I'm looking at if the din rail metal is gone, and the screw is not, can back track? The other pic set attached show the shield block prior to the fix.

1capybara:

For the lighting striking ground up, I think I wasn't clear. I am interpreting when they tell me it can not travel up, they mean that once the potential from the lightning reaches the ground rod, then there is no way it can travel back through the ground wire back up into the panel. I find this false in this situation as if there is a circuit, current will flow. So if the potential is high enough on the ground rod, then yes, it can feed back through on the ground wire, and flow back to the ground rod if a ground loop exist. Ultimately, a single point ground is key as this technical will provide 1 path to the ground as if the potential was the other way from the ground rod, I see this as a open circuit in a single point ground.

 

[Power0020]

Insulation coordination is the process of coordinating the maximum system stress with the minimum insulation strength.

That is by definition, practically, it is to choose equipment based on prospective voltage stresses and specify protective devices accordingly e.g. surge arrestors.

it could have flashed both the terminal and the din rail. the flash will take place at the weakest point due to voltage causing part of surge current to dischrage through the path.

 

[jghrist]

Lightning surges can enter the low voltage system from the ground grid if it is subjected to a lightning strike. The surge has a steep front (equivalent to high frequency) and will cause voltage drops through the inductance of wires throughout the system. The lightning will dissipate through the grid to earth, but because of the steep front it can cause large voltage differences in different parts of the grid. If ground wires are connected to the grid by multiple paths, the surge will take multiple paths through the grounding system and will flash over to parts that are connected to ground in other locations. It is all very complex. Unless you spend a lot of time modeling the system and use a sophisticated computer program that takes the high frequency into account, there is more art than science to surge protection.

Protection can be provided by having low impedance from the grid to earth, and applying surge protectors to limit the voltage between different parts of the system.