bphgravity
Senior Member
- Location
- Florida
The IEC Technical Committee 81 responsible for lightning protection of structures has published hypothetical scenarios for the division of lightning current for structures struck by lightning using both lightning protection, surge protection devices, and employing proper NEC required service grounding. In their report, current distribution through all the connected components of the grounding system should follow Kirchhoff's current law. The NFPA 780 and IEC 81 are based on this principle.
For example, assume two round rods installed roughly 10' apart. One is connected to the LPS the other is the grounding electrode for the building service. Both rods are bonded together. At the service, a SPD is installed. Under the IEC assumption, if 100kA lightning current is injected into the LPS, 50kA would flow to ground in LPS ground rod, 50kA from the LPS rod to the service rod, 25kA to ground through the service rod, 25kA into the house via the GEC, and then 12.5 split between the SPD and the service neutral. Obviously, there would be some variances depending on the true resistances and impedance of the current paths.
In order to test this hypothesis, the above configuration was established at a test house at the International Center for Lightning Research and Testing at Camp Blanding here in Florida. Three tests were performed and the results published in the IEEE Transactions on Power Delivery, Vol. 17, No. 2.
In each of the three tests, the configuration and connections remained the same. Only the ground resistance of each rod was altered for each test. On the third test, the SPD was removed. Attached to the LPI ground rod was a 60' braided tether to a rocket which initiated lightning from natural thunderclouds in order to inject the lighting currents directly to the LPS. The structure was connected to a service entrance form the secondary of a transformer around 150' away but left unenergized.
In test #1, the grounding resistance of each rod was quite high ranging from about 500-ohms to over 1,500 ohms. In test #2 and #3 the ground resistance was reduce to under 100-ohms.
What resulted in quite interesting and unexpected. The ground rods appeared to filter out the higher frequency components of the lightning current, allowing the lower frequency components to enter the house's electrical system. This means that ground rods (electrodes) actually exhibit a capacitive behavior rather than the expected resistive behavior as long assumed. The peak current that entered the home exceeded 80% of the total injected lightning current. There were almost no differences in the measured currents at all points even after the ground resistance was reduced by over 1,000-ohms. Upon excavation of the service cable supplying the structure, 40 pinholes were found in the neutral insulation. On test #3 after the removal of the SPD, significant damage of the service equipment resulted that was not observed after the first 2 tests.
The bottom line is that based on this set of tests a few assumptions can be made:
1. LPS don't work.
2. SPD's do work.
3. Ground rods and other electrodes are useless for lightning currents.
4. The NEC grounding requirements and the NFPA 780 is likely to be ineffective and unfounded.
For example, assume two round rods installed roughly 10' apart. One is connected to the LPS the other is the grounding electrode for the building service. Both rods are bonded together. At the service, a SPD is installed. Under the IEC assumption, if 100kA lightning current is injected into the LPS, 50kA would flow to ground in LPS ground rod, 50kA from the LPS rod to the service rod, 25kA to ground through the service rod, 25kA into the house via the GEC, and then 12.5 split between the SPD and the service neutral. Obviously, there would be some variances depending on the true resistances and impedance of the current paths.
In order to test this hypothesis, the above configuration was established at a test house at the International Center for Lightning Research and Testing at Camp Blanding here in Florida. Three tests were performed and the results published in the IEEE Transactions on Power Delivery, Vol. 17, No. 2.
In each of the three tests, the configuration and connections remained the same. Only the ground resistance of each rod was altered for each test. On the third test, the SPD was removed. Attached to the LPI ground rod was a 60' braided tether to a rocket which initiated lightning from natural thunderclouds in order to inject the lighting currents directly to the LPS. The structure was connected to a service entrance form the secondary of a transformer around 150' away but left unenergized.
In test #1, the grounding resistance of each rod was quite high ranging from about 500-ohms to over 1,500 ohms. In test #2 and #3 the ground resistance was reduce to under 100-ohms.
What resulted in quite interesting and unexpected. The ground rods appeared to filter out the higher frequency components of the lightning current, allowing the lower frequency components to enter the house's electrical system. This means that ground rods (electrodes) actually exhibit a capacitive behavior rather than the expected resistive behavior as long assumed. The peak current that entered the home exceeded 80% of the total injected lightning current. There were almost no differences in the measured currents at all points even after the ground resistance was reduced by over 1,000-ohms. Upon excavation of the service cable supplying the structure, 40 pinholes were found in the neutral insulation. On test #3 after the removal of the SPD, significant damage of the service equipment resulted that was not observed after the first 2 tests.
The bottom line is that based on this set of tests a few assumptions can be made:
1. LPS don't work.
2. SPD's do work.
3. Ground rods and other electrodes are useless for lightning currents.
4. The NEC grounding requirements and the NFPA 780 is likely to be ineffective and unfounded.
