Sub-panels

[weressl]

Lazlo,

It is based on the percent of the total resistance around a 5/8" x 10' ground rod as shown in Chaper 4, Table 9 of the 1991 IEEE Green book. That table shows ~86% of the total resistance within 5' of the ground rod (yes, I said 3' but that was from memory without checking). That would mean if you connect a 10' ground rod to a 120 volt source and touched the source and the earth 5' away from the rod you would receive a 103 volt shock.

Other than lightning protection and possibly votlage stablaziation, there is no reason to connect the system to earth. It does not increase the safety of the system in any way.

They can be a problem on low voltage systems that do not have a code compliant fault clearing path. One example is where a metal parking lot light pole is installed with only a grounding electrode and not the code required equipment grounding conductor. If there is a line to pole fault anyone one touching the pole and the earth 3' away from the pole will receive ~81 volt shock, assuming a 120 volt supply. (Table 9 shows ~68% of the resistance at 3')
Don

If the case is grounded then the short circuit protection will operate. If the case is not grounded then the enclosure will be elevated to line voltage. The totsl circuit resistance will be the body's resistance, including shoewear, and the earth's resistance from the point where the person stands to the nearest connection of the grounding system to the earth.

The total voltage difference is spread along the line between the ground fault point and the earth bonding point of the grounded conductor.

See OSHA 1910.269

""Step potential" is the voltage between the feet of a person standing near an energized grounded object. It is equal to the difference in voltage, given by the voltage distribution curve, between two points at different distances from the "electrode". A person could be at risk of injury during a fault simply by standing near the grounding point."

So if the equipment is grounded there will be no voltage as the OC/SC protective device will open the circuit. If the equipment is not grounded the only time the person would be in danger if the power source AND it's grounded service wire bonding connection to the earth is closer than ~10' or if the person is grabbing a nearby grounded object with his other hand.

The curious can conduct a simple experiment:

Insert a metal plate in your shoe and connect one test-lead of your ohm-meter. Connect your other lead to the grounding electrode. Measure the resistance at 1, 2, 3, 5 and 10' away from the ground rod. Do the math.

 

[iwire]

So if the equipment is grounded there will be no voltage as the OC/SC protective device will open the circuit.

Lazlo, the entire time we are waiting for the OCPD to open the case will be at a elevated potential to ground.

Certainly the hope is the OCPD opens quickly, I can tell you in the real world they do not always open quick.

I have seen dead shorts on long circuits take 4 or 5 seconds to open the breaker.

 

[weressl]

Lazlo, the entire time we are waiting for the OCPD to open the case will be at a elevated potential to ground.

Certainly the hope is the OCPD opens quickly, I can tell you in the real world they do not always open quick.

I have seen dead shorts on long circuits take 4 or 5 seconds to open the breaker.

I agree and that's how they supposed to work. Look at the T/C curves.

If you do not have sufficient fault current due to the high impedance of the long circuit, it will take a while. Hence the GFCI development that will act quickly. But all this have to do with calculated risk. The probability of having a ground fault of a properly constructed installation is very small and the likeliness that the equipment is touched at the same time - during those 3-4 seconds while the SC/OL device opens - infinitestimately small.


GFCI's are being employed in places where the probability of ground fault or ground "leakage" is greatly increased due to special installation circumstances, most often the uncontrollable presence of water.

 

[don_resqcapt19]

Lazlo,

If the case is grounded then the short circuit protection will operate.

If you mean a system that is grounded, has a main bonding jumper and has equipment grounding conductors, I would agree. If there is only a grounding conection to earth and no main bonding jumper, it would be rare for the OCPD to operate for system operating at 600 volts or less.

If the equipment is not grounded the only time the person would be in danger if the power source AND it's grounded service wire bonding connection to the earth is closer than ~10' or if the person is grabbing a nearby grounded object with his other hand.

I have no idea of what that statement means.
Don

 

[George Stolz]

If you do not have sufficient fault current due to the high impedance of the long circuit, it will take a while. Hence the GFCI development that will act quickly.

There is no requirement to GFCI protect light poles, and GFCI protection is not an acceptable substitute for the Equipment Grounding Conductor required to be connected by 410.15.

GFCI's are being employed in places where the probability of ground fault or ground "leakage" is greatly increased due to special installation circumstances, most often the uncontrollable presence of water.

I've installed a fair amount of pole lights (all outdoors), and I've never seen one on a GFCI.

Edit to add: Laszlo, is a grounding electrode required at a light pole served by a single branch circuit?

 

[M. D.]

What is the purpose of separating the grounds & neutrals in a sub-panel ?

because connecting them violates 250.6(A), it would result in a ground loop (parallel neutral current paths) that would allow objectionable current to flow onto metal parts of the electrical installation other grounded metal equipment such as pipes and metal framing

 

[ronaldrc]

Lets assume you did hook your neutral and ground together which defeats the whole purpose of having the separate eq. grd wire and later your neutral came unhooked or loose, you wouldn't even know it untill someone got a good shock out of it.

And if that never happen you would be using your eq. grd as the neutral and wouldn't even know it.