Good Idea or Bad Idea? - High vs Low Current

[breevo]

I recently ran across Mike Holt's YouTube video, "Grounding - Safety Fundamentals (1hr:12min:19sec)" and have learned a lot from this video. I learned the importance of the equipment grounding conductor. I understand that it takes a high current to trip the breaker. But I also learned it doesn't take much current to kill someone. I had a thought though that came to mind recently and would like to know why it is not done this way. Presently, when a fault occurs, the path the electricity should take is the bonded equipment grounding conductor back to the source and with such low impedance, the high current trips the breaker. But you still are relying on a high current. So why don't they have a device right before the equipment grounding conductor is terminated at the source that will trip the breaker if it detects any current (or if you want to be technical, "a current >1 mA")? I would think this would be safer, unless there is a guarantee that all faults will result in a high current at the source. I'm sure there is a reason this is not done, so I'm hoping to learn from your responses. Thanks.

 

[Coppersmith]

If I understand you correctly, we would have one of these devices at every grounding connection in the panel. Essentially a second panel full of very low current breakers would be required. And these breakers would have to shunt trip the circuit breaker controlling the hot. I might work, but it would add a lot of expense and complexity to the system.

 

[Coppersmith]

If every circuit breaker in the panel was a GFCI, you would essentially get this except that GFCI's trip at >30ma.

 

[ActionDave]

I'm sure there is a reason this is not done, so I'm hoping to learn from your responses. Thanks.

We already have a protective device that trips on low current ground faults,
GFCI protection trips at five milli amps and GFPE trips at 30 milli amps.

 

[iwire]

If every circuit breaker in the panel was a GFCI, you would essentially get this except that GFCI's trip at >30ma.

GFCI = Personal protection

Informational Note: Class A ground-fault circuit interrupters
trip when the current to ground is 6 mA or higher and
do not trip when the current to ground is less than 4 mA.
For further information, see UL 943, Standard for Ground-
Fault Circuit Interrupters.

GFP = Equipment protection is often 30 mA

 

[iwire]

So why don't they have a device right before the equipment grounding conductor is terminated at the source that will trip the breaker if it detects any current (or if you want to be technical, "a current >1 mA")?

As Dave mentions we have that device and it is often required.

1 mA would be too low in many cases to deal with normal leakage current in some equipment.

 

[Coppersmith]

GFCI = Personal protection
GFP = Equipment protection is often 30 mA

I stand corrected. :ashamed1: Thanks.

 

[breevo]

Big Thanks for all your replies and thoughts. I still might be missing something here. I thought a GFCI worked differently. You might have to excuse my terminology. I don't deal with this subject on a regular basis. I thought the way a GFCI basically worked was it is continually looking for a difference between the current on the hot lead and the neutral and if this difference is greater than (as I think someone said) 30 mA, it is tripped. This can only occur if the electricity has found an alternative path (i.e. someone getting electrocuted) back to the source. But what I am talking about is a device that senses a current (not a current comparison) on the equipment grounding conductors. If I am understanding this right, all the equipment grounding conductors go to back to the main panel and all terminate a grounding connection, AND there should never be any current on this "network" unless there is a fault. If the device is placed between where all the equipment grounding connections come together and the ground I believe you would only need one device, not one for every equipment grounding conductor. If the device is tripped it would have to send a signal to all the breakers (because it would have no way of knowing which breaker would terminate the fault) or some kind of central remotely controlled main disconnect that would cut the power in front of the breakers. (and yes, these would have to be special breakers in order to trip remotely with the device I'm describing)

Maybe it would just be too expensive and complicated to be practical.

 

[GoldDigger]

Big Thanks for all your replies and thoughts. I still might be missing something here. I thought a GFCI worked differently. You might have to excuse my terminology. I don't deal with this subject on a regular basis. I thought the way a GFCI basically worked was it is continually looking for a difference between the current on the hot lead and the neutral and if this difference is greater than (as I think someone said) 30 mA, it is tripped. This can only occur if the electricity has found an alternative path (i.e. someone getting electrocuted) back to the source. But what I am talking about is a device that senses a current (not a current comparison) on the equipment grounding conductors. If I am understanding this right, all the equipment grounding conductors go to back to the main panel and all terminate a grounding connection, AND there should never be any current on this "network" unless there is a fault. If the device is placed between where all the equipment grounding connections come together and the ground I believe you would only need one device, not one for every equipment grounding conductor. If the device is tripped it would have to send a signal to all the breakers (because it would have no way of knowing which breaker would terminate the fault) or some kind of central remotely controlled main disconnect that would cut the power in front of the breakers. (and yes, these would have to be special breakers in order to trip remotely with the device I'm describing)

Maybe it would just be too expensive and complicated to be practical.

The practical problem here is that there are a lot of conditions that can cause a small leakage current on the EGC, and if you are unlucky these leakage currents can add up to a noticeable current at the common point.
In addition, there are several mechanisms that can lead to high currents through the EGC network even when there are no loads active in the entire building. This is more likely to happen if the ground electrode system does not have a single point of connection to the EGC network.
If you are going to put a detection device on each circuit, the GFCI has the advantage of detecting leakage that goes through a user directly to ground instead of traveling on the EGC.

Your question was a reasonable one, and gives us an opportunity to revisit why GFCIs are useful and what function EGCs perform in both normal and fault conditions.
Thanks for asking.

 

[JFletcher]

Welcome to the forum!

Big Thanks for all your replies and thoughts. I still might be missing something here. I thought a GFCI worked differently. You might have to excuse my terminology. I don't deal with this subject on a regular basis. I thought the way a GFCI basically worked was it is continually looking for a difference between the current on the hot lead and the neutral and if this difference is greater than (as I think someone said) 30 mA, it is tripped. This can only occur if the electricity has found an alternative path (i.e. someone getting electrocuted) back to the source. But what I am talking about is a device that senses a current (not a current comparison) on the equipment grounding conductors. If I am understanding this right, all the equipment grounding conductors go to back to the main panel and all terminate a grounding connection, AND there should never be any current on this "network" unless there is a fault. If the device is placed between where all the equipment grounding connections come together and the ground I believe you would only need one device, not one for every equipment grounding conductor. If the device is tripped it would have to send a signal to all the breakers (because it would have no way of knowing which breaker would terminate the fault) or some kind of central remotely controlled main disconnect that would cut the power in front of the breakers. (and yes, these would have to be special breakers in order to trip remotely with the device I'm describing)

Maybe it would just be too expensive and complicated to be practical.

GFCI breakers and receptacles trip with 6ma or greater differential between current carrying conductors (hot to neutral, hot to hot).

What you are describing is basically a panel main breaker with GFCI protection.

Also, if a fault occurs, the EGC may not see any current from that fault. A GFCI receptacle would, e.g.: a kid sticking a pin in a non TR GFCI receptacle. Current flows thru the kid to a lower potential object, like his hand on a metal water pipe or radiator. The EGC sees no current yet the GFCI trips because no (or less) current is returning on the grounded conductor than what is going out on the ungrounded conductor.

 

[breevo]

Thank you. I appreciate all your replies. This has been very helpful.

 

[brian john]

What you are describing is what is commonly referred to as Ground Return GFPE and was used in early 480/277 VAC switchboards for Ground Fault Protection. This system is also common in medium voltage systems for GFPE protection and some Main Tie Main switchboards.

The problem with what you describe, as I believe others have noted, is all the fault current may not be on the EGC (Equipment Grounding Conductor). The only way this could work with some level of certainty would be to put your CT on the Main Service Bonding screw or jumper. Then the issue becomes what level do you set the relay or CB at, as was noted leakage current can be well above your suggested 1ma, when you have a large residence it may be possible to have currents above your 1ma and particularly in large facilities it is not uncommon to see several amps.

Additionally in large commercial office buildings we find branch circuit neutrals with downstream grounds very common this current is quite often in the 10-20 amp range when measured at the main bonding jumper.

Lastly in residence and commercial buildings that share a common metallic water or gas piping, it is not uncommon to see circulating ground current from the pipes being in parallel with the neutrals to each structure.

Additionally the existing GFCI basically offer the same protection your are describing only in a more reliable way.