Ground Impedance

[iwire]

Much like the wheels on the tracks of the old time electric train sets.

JAP>

:thumbsup:

 

[tom baker]

Even with good low impedance ground you might see ground loops which will cause problems on such systems if not properly designed.

-Jon

I have heard this comment before, can you explain what a ground loop is and what problems it can cause?

 

[charlie b]

The Code generally requires a "low impedance" path to ground. . . .

Please allow me to split a hair here. No, the code does not require that. Specifically, the required low impedance path is the path that returns current to its source. The “ground” is not a player in this discussion.

I am involved in a situation where existing equipment grounding conductors (via metallic raceways) is, worst case, about 1/10 of an ohm.

120 volts divided by 0.1 ohms gives a fault current of 1,200 amps. That is 60 times the rated current of a 20 amp breaker. That is very likely to put you into the instantaneous region of the breaker. But that is one example. If you have a 40 amp breaker, that might not put you into the instantaneous region. Some engineering is needed to establish the actual operating conditions of your installation.

The grounding electrode impedance is 2.5 ohms, so that is not in question

More to the point, it is not relevant at all. As I said earlier, the ground is not a player. If there is a short circuit from somewhere internal to a load to the external metal case of the load, then the EGC (be it a conduit or a wire-type) will carry a large amount of fault current back to the source. That current will not flow into the ground via the ground rod. Rather, it will flow from source to load to fault point to external metal case to EGC and back again to the source, and it will continue to do so until the overcurrent device opens to terminate the event.

 

[jap]

I'd have to disagree that absolutely no fault current will flow into the ground through the ground rod during a fault on the loadside of a service transformer.
I'd tend to believe that during a fault the fault current tries to get back to its source through any and all avenues.

JAP>

 

[jap]

Maybe I misunderstood you.

JAP>

 

[winnie]

I have heard this comment before, can you explain what a ground loop is and what problems it can cause?

The term ground loop is usually applied to noise pickup by unbalanced signal lines, where you have one signal conductor and a grounded 'shield' conductor. If you have multiple 'grounded' paths between pieces of equipment, those closed loops can act as transformer secondaries, pickup up alternating magnetic fields surrounding the equipment and coupling this to your signal equipment.

I guess I am mis-applying the term to a similar situation in power distribution.

Code requires you to arrange your circuit conductors so that all of the conductors in a conduit taken together have zero net current. This means that current flowing on one conductor is supposed to be balanced by the current flowing in the other conductors. (There are exceptions, eg. in non-metallic raceways...)

If you do have net current in a conduit, then the magnetic field produced by the conductors can induce voltage in the conduit.

If, because of the conduit arrangement you have a _loop_ of conduit, then the induced voltage will cause current to flow in the conduit.

In essence, the circuit conductors are acting as the primary of a transformer, and the conduit is acting as a shorted secondary. Not a very efficient transformer, and the current in the secondary is limited by the current in the primary. But if the current flow is high enough the conduit can get pretty toasty.

I am pretty sure that iwire once posted a picture of a service that failed. It was installed in metallic conduits with isolated phase conductors (phase A in one conduit, phase B in the next...). I tried but can't find the post; not sure if this is a false memory

-Jon

 

[winnie]

I'd have to disagree that absolutely no fault current will flow into the ground through the ground rod during a fault on the loadside of a service transformer.
I'd tend to believe that during a fault the fault current tries to get back to its source through any and all avenues.

JAP>

charlie said that the ground is not a player...but he really should have said that the ground is a negligible player

With any reasonable bonding arrangement and any reasonable ground, a very small fraction of the current will return to the source via the soil and the vast bulk of the current will return via the metallic EGC system and the grounded conductor between the service and the transformer.

There are exceptions, eg common metallic water piping used as ground, but these exceptions make the point that current flow through the soil is negligible for fault current on low voltage systems.

-Jon

 

[kwired]

I'd bet Its not static.
The person taking off the lid is becoming part of the return path himself, its just that he has too much resistance feel a noticeable shock.

JAP>

If it makes a spark at 120 volts when you break the circuit it will be enough current you are going to feel it, if it is static it can be tens of thousands of volts, which is why it can spark, but the current level is so low it isn't usually very harmful, but sometimes can be.

 

[jap]

charlie said that the ground is not a player...but he really should have said that the ground is a negligible player

With any reasonable bonding arrangement and any reasonable ground, a very small fraction of the current will return to the source via the soil and the vast bulk of the current will return via the metallic EGC system and the grounded conductor between the service and the transformer.

There are exceptions, eg common metallic water piping used as ground, but these exceptions make the point that current flow through the soil is negligible for fault current on low voltage systems.

-Jon

Well put.

JAP>

 

[jap]

If it makes a spark at 120 volts when you break the circuit it will be enough current you are going to feel it, if it is static it can be tens of thousands of volts, which is why it can spark, but the current level is so low it isn't usually very harmful, but sometimes can be.

i don't know that I'd call it breaking of the circuit as much as I'd describe it as the creation of a path to a grounded conductor that was creating the continual shower of sparks from the end of the conduit I was holding,to the conduit run along the walls of the business,as I was running it back and forth, but we never got shocked.

JAP.

 

[JFletcher]

If you have a 40 amp breaker, that might not put you into the instantaneous region.

Im no engineer, however if 600A doesnt trip a 40A breaker within a fraction of a second, I'd call that seriously faulty. or am I being unrealistic? Would an average 40A breaker hold 600A for 10Hz? 2Hz? Even at 120V that would be 0.2ohms resistance, which is close enough to a bolted fault.

 

[winnie]

Im no engineer, however if 600A doesnt trip a 40A breaker within a fraction of a second, I'd call that seriously faulty. or am I being unrealistic?

Take a look at a breaker trip curve. For example http://static.schneider-electric.us...it Breakers/QO-QOB Circuit Breakers/730-4.pdf

At 10x the handle current, the breaker might hold for up to 0.8 seconds (the top of the trip curve), or it might trip 'instantaneously' (the bottom of the trip curve). The breaker is specified to trip somewhere in that range, and anywhere in that range is considered acceptable.

-Jon

 

[kwired]

i don't know that I'd call it breaking of the circuit as much as I'd describe it as the creation of a path to a grounded conductor that was creating the continual shower of sparks from the end of the conduit I was holding,to the conduit run along the walls of the business,as I was running it back and forth, but we never got shocked.

JAP.

After that description, sounds like current is flowing in your raceways (maybe a neutral bonded to them somewhere). You are making and breaking a parallel component, significant enough current is flowing it creates a spark, but open circuit voltage when you disconnect that segment is low enough you don't feel anything. You did not interrupt the entire circuit, just opened one parallel component. If there were just two paths of equal resistance you broke half the total current.

 

[jap]

After that description, sounds like current is flowing in your raceways (maybe a neutral bonded to them somewhere). You are making and breaking a parallel component, significant enough current is flowing it creates a spark, but open circuit voltage when you disconnect that segment is low enough you don't feel anything. You did not interrupt the entire circuit, just opened one parallel component. If there were just two paths of equal resistance you broke half the total current.

If a Neutral conductor at the service was too small to carry all of the neutral return current would it contribute to this issue?


JAP>

 

[kwired]

If a Neutral conductor at the service was too small to carry all of the neutral return current would it contribute to this issue?


JAP>

If the alternate path were less resistance then the undersized neutral then more current would flow on the alternate path then through the neutral when in parallel.

 

[charlie b]

I'm no engineer, however if 600A doesn't trip a 40A breaker within a fraction of a second, I'd call that seriously faulty.

I merely stated that that trip might not take place within the instantaneous region.

Would an average 40A breaker hold 600A for 10Hz? 2Hz?

Looking at the example TCC curve to which Jon posted a link, my answers are no and yes, in that order. 10 cycles (not Hz, by the way, as that is cycles per second) is equal to .167 seconds, and the breaker trip curve does not go that high for a fault of 15 times the breaker's rated current. 2 cycles is equal to .033 seconds, that lies within the breaker trip curve's range.

 

[cpuerto]

Ground Impedance

Just ohm law. If r=0.1 ohm and the voltage is 120 V, then the current is 120/0.1=1200 A. If your breaker is 1200 A, most probably never will trip. If you breaker is 200 A, maybe it will take a minute or so to trip. No really a safe system. So everything depends of the voltage and the tripping curve of the breaker an the time to clear the fault.