- Location
- Illinois
- Occupation
- retired electrician
We made second building installs like that for a long long time without out any real world issues. I think the 2008 code took that away.
MBJ Location
- Thread starter dereckbc
- Start date
- Status
- Occupation
- Licensed Electrician
First, I don't see how that is possible because there is a grounded conductor from the pole to the shelter.
Second, the fix that the OP want to implement involves nothing more than adding a screw to the power panel at the shelter. All the wiring stays the same.
How is it possible that a metal screw, smaller than the size of a baby's pinkie finger can change a radio tower from one that blows up equipment during a lightning event into one where everything stays safe? It is beyond my understanding.
I will try to help. The grounded circuit conductor (service neutral) is bonded to a pole butt ground 100 cable feet away from the building, 80 feet walking distance. The Neutrals ZERO-VOLT Reference Point is at the service pole, and the buildings Zero-Volt is at the building GES . Once the service enters the building, the neutral is not bonded. The building has its own Ground Electrode System, and it is not bonded to the Pole Butt Ground where the neutral is bonded. That in itself is a code violation. The two earth grounds are at different voltage potentials.
Under normal operating conditions the voltage difference between the two points is not great enough to be an issue. The only difference you see is the Current * Resistance voltage drop developed on the neutral as it travels downstream. Normal voltage drop one expects as load current increases.
The problem manifests itself when there is a lightning strike. When lightning strikes and enters the dirt, there is a significant voltage gradient along the surface of the ground. The neutral is at the potential of the pole butt ground, while the building's GES is at another potential. The difference between N-G voltage at the building entrance is in the 10's of thousands of volts. That potential difference is felt at all equipment with a Neutral circuit or wherever they come in close proximity. It flashes over.
Contrast that to bonding the Neutral and earth ground where it enters the building. Does not matter what happens upstream, the voltage will always be 0-volts between N-G. The ground potential rises, but with a single point ground, no potential difference is felt inside. Like a fishing bobber on a lake floats above the waves. It does not know the waves are going up and down.
Perhaps you have heard of Stray Voltages or Step Potential? Both are the exact same thing. Stray voltages are experienced by farmers and to some extent city street lights. For example a livestock water tub heater in a cow pasture with the electric service some distance away. The tank is bonded to the ACEG which is a referenced to earth some distance away. A cow walks up standing in the mud and tongue touches the water and receives a mild shock. Just enough the animal does not drink. Dairy farms have a problem with Stray Voltage milking machines.
Step Potential is the same thing on Steriods and the steroid is lightning. As lightning enters the earth spreads out across the surface causing a very significant voltage gradient. About 100-volts per inch of earth on average. The distance between your feet is enough to kill and roast chestnuts and family jewels. If you do a google search on 71 people killed in India will get you a lot of news reports because 71 people were killed in Inia 2-weeks ago from 4 different lightning events. All involved hitting a tower or structure. Step Potential killed all the victims.
Hope that helps.
Last edited:
Either 2005 or 2008. So I have 3 code issues to work with. 90.4 waiver should work because it is reviewed and approved by a PE. 250.6 (B), and I think 250.94 applies.
What do you think of 250.94? Besides we are a utility, rules do not apply to us
Last edited:
- Occupation
- Licensed Electrician
It would help if you answered my question about the wiring between the pole and the shelter. If I understand correctly you only have three wires, two hots and a grounded/neutral conductors between the shelter and the pole and an unbonded neutral at the shelter.I will try to help. The grounded circuit conductor (service neutral) is bonded to a pole butt ground 100 cable feet away from the building, 80 feet walking distance. The Neutrals ZERO-VOLT Reference Point is at the service pole, and the buildings Zero-Volt is at the building GES . Once the service enters the building, the neutral is not bonded. The building has its own Ground Electrode System, and it is not bonded to the Pole Butt Ground where the neutral is bonded. That in itself is a code violation. The two earth grounds are at different voltage potentials.
Under normal operating conditions the voltage difference between the two points is not great enough to be an issue. The only difference you see is the Current * Resistance voltage drop developed on the neutral as it travels downstream. Normal voltage drop one expects as load current increases.
The problem manifests itself when there is a lightning strike. When lightning strikes and enters the dirt, there is a significant voltage gradient along the surface of the ground. The neutral is at the potential of the pole butt ground, while the building's GES is at another potential. The difference between N-G voltage at the building entrance is in the 10's of thousands of volts. That potential difference is felt at all equipment with a Neutral circuit or wherever they come in close proximity. It flashes over.
Contrast that to bonding the Neutral and earth ground where it enters the building. Does not matter what happens upstream, the voltage will always be 0-volts between N-G. The ground potential rises, but with a single point ground, no potential difference is felt inside. Like a fishing bobber on a lake floats above the waves. It does not know the waves are going up and down.
Perhaps you have heard of Stray Voltages or Step Potential? Both are the exact same thing. Stray voltages are experienced by farmers and to some extent city street lights. For example a livestock water tub heater in a cow pasture with the electric service some distance away. The tank is bonded to the ACEG which is a referenced to earth some distance away. A cow walks up standing in the mud and tongue touches the water and receives a mild shock. Just enough the animal does not drink. Dairy farms have a problem with Stray Voltage milking machines.
Step Potential is the same thing on Steriods and the steroid is lightning. As lightning enters the earth spreads out across the surface causing a very significant voltage gradient. About 100-volts per inch of earth on average. The distance between your feet is enough to kill and roast chestnuts and family jewels. If you do a google search on 71 people killed in India will get you a lot of news reports because 71 people were killed in Inia 2-weeks ago from 4 different lightning events. All involved hitting a tower or structure. Step Potential killed all the victims.
Hope that helps.
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
No, he has 4 wires between the shelter and pole, with the MBJ at the pole. 3 wires with an unbonded neutral at the shelter would be untenable, as there would be no fault clearing path for the bonding conductor.
Cheers, Wayne
- Occupation
- Licensed Electrician
I know. But I see no mention of a fourth wire except for the part where he talks about adding one.
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
- Occupation
- Licensed Electrician
Ok. I don't know how I missed that.
Since there is an EGC connected to the electrical equipment at the shelter and it goes back to the utility pole then the grounding electrode system at the shelter and the but ground at the pole are only geographically isolated and not electrically isolated. Am I understanding this correctly?
- Location
- Illinois
- Occupation
- retired electrician
I believe that 250.94 was added to try to avoid the exact issue you are trying to fix in your installation. There were cases where the the communications utility had their own grounding electrode on the other side of the building from the electrical grounding electrode.Either 2005 or 2008. So I have 3 code issues to work with. 90.4 waiver should work because it is reviewed and approved by a PE. 250.6 (B), and I think 250.94 applies.
What do you think of 250.94? Besides we are a utility, rules do not apply to us
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
Not sure as I lost you there. Within the shelter, there's an EGC that's connected to the shelter GES and the outdoor feeder EGC and thus to the pole GES. There's also a grounded conductor (neutral) that's not connected to the shelter GES but is connected to the pole GES.
So when there's a voltage difference between the pole GES and the shelter GES, I expect that current must flow on the outdoor feeder EGC. I guess the shelter EGC will be near the potential of the shelter GES. While the grounded conductor potential will be closer to the pole GES potential, as it's not connected to the shelter GES. That grounded conductor / EGC potential difference is the problem. [And maybe the presence of a DC system one of whose conductors is connected to the shelter GES compounds the problem, as that DC conductor may be near the AC grounded conductor.]
If the grounded conductor is rebonded at the shelter to the EGC and shelter GES, then instead the outdoor grounded conductor and the outdoor EGC will both carry current from the difference in earth potential at the two GESs. But the grounded conductor in the shelter should be closer to the shelter GES potential, reducing or eliminating the grounded conductor/EGC potential difference.
Cheers, Wayne
Yes sir that is correct. The system is not safe as configured. A simple bond at the building entrances corrects the issue.
- Occupation
- Licensed Electrician
Yes. Agreed and understood.
Yes.
And here is where my understanding starts to break down. The pole GES and shelter GES are connected to each other, there is an equipment grounding conductor that is connected to all the metal parts including the pole GES going over to the shelter and connecting all the metal parts including the shelter GES.
Or to put it another way, pre 2005 the grounded/neutral conductor was allowed to double as the equipment grounding conductor, post 2005 the two are separate, but electrically there is no difference.
And that rebonding of the neutral usually involves using a screw inside the electrical disconnect, but we haven't taken two isolated GESs and connected them, we've just added a screw.
- Occupation
- Licensed Electrician
Don't give up on me. I'm gonna get this clear in my head one way or another.
Correct you are. Telephone, Antenna Installers, SATV, and CATV were notorious for doing that. Ham radio operators are the biggest threat today. They bring their coaxes or cable to the premises at a different location than the AC Service enters. Drive a ground rod where their service or coax enters, bond their service to it, and walk away.
Those service providers had to pay a lot of damages to equipment and some house fires with that practice. It placed their homes and equipment in a nasty nasty ground loop between two earth grounds. The house wiring and equipment were used as a bonding jumper between two grounds.
So thanks Don I think I have my answer and code references. 90.4, 250.6, and 250.94.
Any others that might be helpful?
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
Yes. But the feeder EGC connection doesn't guarantee that the shelter GES and pole GES are always at the same potential. Rather, when the lightning induces a voltage difference between earth at the pole and earth at the shelter, that EGC connecting the pole GES and the shelter GES is going to have an induced current on it.
E.g. (to pick some inaccurate made up numbers), let's say that there's a 1000V difference; call the shelter GES 0V and the pole GES 1000V. If the outdoor EGC has an impedance of 1 ohm for the lightning waveform, then 1000A will flow on the outdoor EGC (at least temporarily). The pole GES stays at 1000V.
Now without the shelter grounded conductor-GES bond, the grounded conductor does not have any current flowing on it from the intentional connections. It will try to stay at 1000V. But where it is routed close to the shelter EGC or bonded metal components, which are at 0V, that arrangement is not likely to have been designed to withstand 1000V. You'll have an arc at some unplanned location, causing some current to flow on the grounded conductor and damage to the components involved.
If, instead, at the point the grounded conductor enters the shelter, the grounded conductor is rebonded to the shelter GES, then the outdoor grounded conductor will be carrying current through that connection, just like the EGC does. So if its impedance is 1 ohm, another 1000A will flow on the grounded conductor (at least temporarily). That makes the potential of the shelter end of the grounded conductor 0V now, instead of 1000V. This eliminates (approximately) the grounded conductor - EGC potential difference inside the shelter, so you don't get random arcing in equipment.
Cheers, Wayne
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
To say the same thing perhaps more succinctly, when lightning induces a potential difference between the two GESs, they behave like a power source. The outdoor EGC is just a dead short across the two sides of that power source. But the indoor EGC is connected to one side of that power source, and a non-rebonded grounded conductor is connected to the other side of that power source. Whereas rebonding the grounded conductor at the shelter turns the outdoor portion of the grounded conductor in another dead short, while making the indoor grounded conductor and the indoor EGC both connected to the same side of that power source.
Cheers, Wayne
Cheers, Wayne
- Location
- Illinois
- Occupation
- retired electrician
I think those cover the issue, Dereck.
Nah, I will drag you kicking and screaming.
I teach protective bonding to a lot of engineers, technicians, and sparkies. Part of that class involves busting a few electrical myths that prevent us from understanding. The very first one you might be hung up on is: "Ground rods are at zero-volt potential with respect to each other."
That myth gets many folks into trouble and is the root cause of damage from lightning and high voltage utility faults. Drive a 10-foot ground rod into the ground, drive another 20 feet away, and there will be a difference of potential between them. . So imagine the diagram below. AC Service enters one side of the home, and hammy sammy brings in the coax on another side. Sammy drives a rod outside where his coax enters to bond his coax shield and ADU.
Look at the resulting circuit we built. Two ground rods stuck in the mud some distance apart with your electronics and house wiring acting as a bonding jumper between the two earth grounds. That is a Ground Loop, a nasty ground loop. If lightning were to strike either the antenna, a nearby structure or arrive via utility, it would place those two ground rods at thousands of volts potential between them. That leaves your bonding jumper the job of equalizing the two points. It would be the equivalent of taking, say, a 13.2 kV utility feeder phase A to one ground rod and phase B to the other rod. Your 12 AWG 20-amp branch circuit ACEG wire and radio coax are bonded together in the radio equipment to complete the circuit path. You provided a path for lightning to enter and exit right through your home. With thousands of amp flowing vaporizes the coax shield and ACEG wire while having a look around inside your radio equipment.
It is for this very reason NEC added 250.94, single-point ground. It came from the wireless/telecom industry, Bell Labs, and the Lightning Protection Institute. Having a massive GES like wireless and telecom is extremely useful and effective if used properly. The trick is to bond the GES one time only at a single point. Once a cable enters, it never sees earth ground again and is kept isolated. Bonding jumpers are single-wire-circuits and are an open circuit so no current can flow through them. Just a wire coming inside that goes nowhere electrically with one end stuck in the mud outside. Holds everything at the same potential and provides a return fault path for interior faults while keeping external faults outside where they belong.
- Status